research.library.mun.caall fetal hean rate and uterine activity graphs were classified according to...
TRANSCRIPT
1'*'1 National Library of canada
Bibliotheque nationals duCanada
AcQuisitions and Bibliographic ServiceS
Acquisitions et services bibliographiques
385 Wllinglon SlrNt oaawa ON K1A ON4 CaMda
385, rue Wellington Oltawa ON K1 A ON4 canada
The author bas granted a nonexclusive licence allowing the National Library of Canada to reproduce, lo311y distribute or sell copies of this thesis in microform, paper or electronic formats.
The author retains ownership of the copyright in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.
L' auteur a accorde une licence non exclusive permettant a Ia Bibliotheque nationale du Canada de reproduire, preter, distribuer ou vendre des copies de cette these sous Ia forme de microfiche/film, de reproduction sur papier ou sur format electronique.
L' auteur conserve Ia propriete du droit d'auteur qui protege cette these. Ni Ia these ni des extraits substantiels de ceUe-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation.
0-612-62369-6
Canadl
A MASKED RANDOMIZED COMPARISON OF
ORAL AND VAGINAL ADMINISTRATION OF
MISOPROSTOL FOR LABOUR INDUCTION
by
Kelly Angela Bennett
A thesis submitted in partial ful6llment of the requirements for the degree of
Master of Science
Memorial University of Newfoundland
2000
A MASKED RANDOMIZED COMPARISON OF
ORAL AND VAGINAL ADMINISTRATION OF
MISOPROSTOL FOR LABOUR INDUCTION
by
Kelly Angela Bennett
A thesis submitted in panial fulfillment of the requirements for the degree of
Master of Science
Memorial University of Newfoundland
2000
Approved by -----~~=-------=-::~-~-=---=---Chairperson of Supervisory Committee
Program Authorized toelff~D~~-----------------------------------------
Date -----------------------------------------------
ABSTRACT
Objective: To test the null hypothesis that administering misoprostol orally or vaginally will
result in no difference in time to vaginal birth, and to determine whether different frequencies
of tachysystole and hyperstimulation are associated with route of administration.
Methods: Two hundred six women after 37 completed weeks' gestation who presented with an
indication for induction were randomly assigned to receive misoprostol (50 fJg) orally or
vaginally every 4 hours as needed to induce labour. Placebo use and-allocation concealment
accomplished blinding until data analysis was completed. Sample size was calculated to allow a
two-tailed a of .05 and power (1-~) of 80%. All fetal hean rate and uterine activity graphs
were classified according to Curtis' criteria before the induction groups were unmasked.
Results: Analysis involved 104 women in the oral group and 102 in the vaginal group. The
mean time(± standard deviation) to vaginal birth with oral misoprostol was 1072 (± 593)
minutes compared with 846 (± 385) minutes with the vaginal protocol (P == .004). There were
no significant differences in cesarean rate, epidural use or neonatal outcomes. More frequent
tachysystole for 20 minutes (P < .01) and hyperstimulation (P < .04) were observed with
vaginal misoprostol. No neonatal asphyxia occurred in either group.
Cond~tsion: Misoprostol effectively induces labour, given orally or vaginally. There is a shorter
interval to vaginal birth with vaginal application; however, the more frequent occurrence of
fetal heart graph abnormalities in this group suggests that, until the optimal dosing interval for
vaginal use is determined, the preferred route of misoprostol administration might be oraL
2
DEDICATION
For my parents.
3
ACKNOWLEDGMENTS
The author wishes to acknowledge support provided in the form of a research grant
from the Health Care Corporation of St.John's and Memorial University of Newfoundland,
Faculty of Medicine, St.John's, Newfoundland, Canada. The author gratefully acknowledges
the assistance of the following individuals in the preparation, planning and execution of this
project.
• Dr. David C. Young
• Dr. Joan M. G. Crane
• Dr. Kimberley Butt
• Dr. Rory Windrim
• Dr. George Fox
• Dr. John Fardy
• Dr. Stephane A. Braiin
• Ms. Donna Hutchens - Research Nurse
• Mr. Jerry Peckham - Pharmacist
• Obstetricians, Residents and Labour & Delivery Nurses of the Grace General Hospital
4
TABLE OF CONTENTS
ABSTRACf ............................................................................................................................. 2
DEDICATION ......................................................................................................................... J
ACKN'OWLEDGMENTS ..................................................................................................... 4
TABLE OF CONTENTS ....................................................................................................... 5
LIST OF TABLES ................................................................................................................ 1 0
LIST OF FIGU'RES .............................................................................................................. 1 1
CHAPTER 1 ............•.............................................................................................................. 12
Introduction ................ ..................................................................................................... I 2
lnduction of Labour .................................................................................................... 12
Cervical Ripening for Labour lnduction .................................................................... 13
CHAPTER 2 ........................................................................................................................... 15
The Role of Prostaglanciins In Labour ........................................................................... / j
Prostaglandin Synthesis .............................................................................................. 15
Prostaglandin Receptors ............................................................................................. 1 7
Prostaglandins and Parturition ................................................................................... 18
Prostaglandins and Cervical Ripening ....................................................................... 20
CHAPTER 3 .................................................................. ··············································· .......... 22
Mechanical Methods of Labour lnduction ..................................................................... 22
5
Sweeping Membranes ................................................................................................ 22
Balloon Catheters ....................................................................................................... 24
Synthetic Hygroscopic Dilators ................................................................................. 26
Natural Hygroscopic Dilators .................................................................................... 27
CHAPTER 4 ........................................................................................................................... 29
Pharmacologic Induction Agellls ................................................................................... 29
Phannacological Induction Agents ............................................................................ 29
Dinoprostone ............................................................................................................... 30
Controlled-Release Insert ........................................................................................... 31
Misoprostol ................................................................................................................. 33
Review of Vaginal Misoprostol Literature ................................................................ 33
Review of Oral Misoprostol Literature .................................................................. 44
CHAPTER 5 ........................................................................................................................... 48
Research Question and Methocls .................................................................................... 48
Research Question ...................................................................................................... 48
Scientific Rationale and Originality .......................................................................... .48
Research Design ......................................................................................................... 49
Sample Size Calculation ............................................................................................. 49
Setting ......................................................................................................................... 49
Inclusion and Exclusion Criteria ................................................................................ SO
Randomization ............................................................................................................ SO
Allocation Concealment ............................................................................................. 51
6
Intervention ................................................................................................................. 51
C l. -., omp 1ance ................................................................................................................. :> _
Primary Outcome ........................................................................................................ 52
Secondary Outcomes .................................................................................................. 33
Data Analysis .............................................................................................................. 54
Confidentiality and Ethical Issues .............................................................................. 54
Study Protocol for Misoprostol Induction .............................................. .................. 55
Contraindications .................................................................................................... 55
Note ......................................................................................................................... 56
Procedure ................................................................................................................ 56
Budget and Grant Applications .................................................................................. 57
CHAPTER 6 ........................................................................................................................... 59
Results ........................................................................................... .................................. 59
Study Population ......................................................................................................... 59
Primary Outcome ........................................................................................................ 60
Secondary Outcomes .................................................................................................. 63
Birth Route ............................................................................................................. 63
Excessive Uterine Activity ..................................................................................... 63
Neonatal Outcomes ................................................................................................ 65
Labour Satisfaction ................................................................................................. 67
CHAP'fER 7 ........................................................................................................................... 68
Methodological Issues and Discussioll .......................................................................... 68
7
Design Problems ......................................................................................................... 68
Recruitment ................................................................................................................. 70
Classification of Excessive Uterine Activity ............................................................. 71
Pharmacokinetics of Crushed Misoprostol ................................................................ 72
A Future Study of Excessive Uterine Activity .......................................................... 73
Study Design .......................................................................................................... 73
Methods .................................................................................................................. 74
Sample Size Calculation ........................................................................................ 75
CHAP'fER 8 ........................................................................................................................... 76
Is Oral Misoprostol The Idea/Induction Agell/? ........................................................... 76
Misoprostol ................................................................................................................. 76
Pharmacokinetics of Oral and Vaginal Misoprostol ............................. .................... 77
Misoprostol-Current and Future Use ......................................................................... 79
RE FERENCES ...................................................................................................................... 82
APPENDIX A: PEER-REVIEWED JOURNAL PUBLICATION ••••••••••••••.••••.•••••••••••• 9S
APPENDIX 8: HUMAN INVESTIGATION COMMITTEE APPLICATION FORM
•.••••••••••.•••••••••••••••••••••••••••••••••••••••••••••••...••••••••••••••••.•••.•.••••••..•••••.••.•••••••.•.••.••.••.•....•.•••..••....•• 101
APPENDIX C: OAT A COLLEcriON FORMS .......................................................... I 06
APPENDIX D: POSTPARTUM MISOPROSTOL QUESTIONNAIRE •••••••.•••••••••• 109
APPENDIX E: CONSENT FORM ................................................................................. 112
8
APPENDIX F: MEDICAL RESEARCH COUNCIL GRANT APPLICATION ..••• liS
APPENDIX G: BUDGET ................................................................................................. 136
9
LIST OF TABLES
M~ fu~
Table 1: Summary of Randomized Trials of Vaginal Misoprostol for Labour
Induction .............................................................................................................. 4 3
Table 2: Summary of Randomized Trials of Oral Misoprostol for Labour
Induction .............................................................................................................. 4 7
Table 3: Demographic characteristics .......................................................................... 61
Table 4: Peripatt\Jlll data ................................................................................................ 62
Table 5: Labour inteiVal to vaginal binh ..................................................................... 63
Table 6: dassification of excessive uterine activity ................................................... 64
Table 7: Neonatal outcome ........................................................................................... 66
10
LIST OF FIGURES
NIIITih!r Page
Figure 1: Stuey population ........................................................................................... 60
ll
Chapter 1
INTRODUCTION
Induction of Labour
Induction of labour is the initiation of cervical ripening and uterine contractions
before their spontaneous onset. Induction is often indicated for a variety of obstetrical,
medical or fetal complications of pregnancy. Various mechanical and pharmacological
methods are currently used to induce labour; however, no single method or agent has been
suitable for all clinical situations. All available methods of induction of labour have associated
medical risks; therefore, the decision to induce labour should only be made when the risk of
continuing a pregnancy outweighs the risk of induction.
For cenain clinical conditions such as severe preeclampsia, the decision to induce
labour is straightforward. For other conditions, the point at which the risk of continuing
pregnancy outweighs the risks associated with induction is not clear. Funhermore, the risk-to
benefit ratio may be influenced by the methods used to induce labour, and by the methods
used to assess fetal well-being when expectant management is chosen. Induction requires
continuous electronic fetal and uterine contraction monitoring, which often reduces a woman's
mobility md comfon.
Ideally, induction agents should mimic spontaneous labour while avoiding e.""<cessive
uterine activity. However, because the mechanisms that control the initiation of parturition are
12
not well understood, such goals are difficult to achieve. The major concerns associated with
induction are ineffective labour and excessive uterine activity. Both problems may lead to an
increased risk of cesarean delivery. Excessive uterine activity is defined as contractions of
abnormally high intensity or frequency that lead to impaired uteroplacental circulation and
consequently, decreased fetal oxygenation. When contractions of abnormally high intensity or
frequency occur, insufficient recovery time resulting in fetal heart rate abnormalities could
necessitate operative delivery or lead to fetal asphyxia.
Cervical Ripening for Labour Induction
In 1964, Bishop1 described a pelvic scoring system that predicted vaginal delivery in
multiparous women. The five components of the score were: position, consistency,
effacement and dilatation of the cervi.x and station of the fetal venex. A ma.ximum score of
thineen described a ripe cervix. Bishop noted that with a score of nine or more the risk of
failed induction was minimal. Bishop scores of 0-3 indicated a high likelihood of failed
induction. This served as the basis for the concept of ripening the cervi.x for women whose
score is unfavourable, before attempting induction with oxytocin.
Since the discovery of prostaglandins and their use as agents for induction of labour,
a wide variety of different formulations, routes of administration and dosage regimens have
been investigated. Currently, prostaglandin PGE! (dinoprostone) is the most widely used for
the purpose of induction, with the preferred route of administration per vagina. A variety of
vehicles have been developed for the delivery of this agent, including lactic acid-based
pessaries, water-soluble gels and most recently, a slow -release hydrogel polymer.
13
Cbapter 2
THE ROLE OF PROSTAGLANDINS IN LABOUR
Prostaglandins are lipid molecules that act as intermediaries in several physiologic and
pathologic processes. Prostaglandins of theE (PGE1, PGE~ and F (PGF ~) series have potent
uterotonic effects and are commonly used as labour induction and cervical ripening agents.
Recent advances in molecular biology and biochemical techniques have allowed deeper
understanding of the role prostaglandins play in the process of parturition.
Prostaglandin Synthesis
Biologically active eicosanoids are formed from polyunsaturated fatty acids.
Arachidonic acid is the most common naturally occurring precursor and must be released
from cell membrane stores, where it exists in an esterified form. The liberation of arachidonic
acid from membrane phospholipids is generally considered the initial step in the synthesis of
prostaglandins.~ This release from the esterified position is accomplished by the action of
phospholipases, primarily phospholipase A2 (PLA~. PLA2 catalyzes the liberation of
arachidonic acid from phosphatidylethanolamine, phosphatidylserine and phosphatidylysine . .!
PLA2-type IT is the most abundant of four subtypes of the enzyme in the placenta, fetal
membranes and decidua. The activity ofPLA2 is increased in the amnion and placenta during
labour.3 The action of phospholipase C (PLC) is indirect, catalyzing the hydrolysis of
15
Oral administration of prostaglandins in obstetrics has virtually been abandoned
because of the severe gastrointestinal side effects associated with existing preparations.
Administration of prostaglandins for induction of labour has been limited to intracervical and
intravaginal routes. An oral induction agent that is safe, effective, inexpensive and well
tolerated by patients would likely be attractive to patients and health care providers. Several
researchers have conducted randomized trials comparing vaginal misoprostol to standard
therapy and have found it to be a safe and effective prostaglandin for induction of labour. No
previous masked randomized comparisons of oral and vaginal misoprostol (SOJ.1g every four
hours) for labour induction have been published in the English language literature.
14
phosphatidylinositol into diacylglycerol and inositoltriphosphate. Diacylglycerol is then
metabolized by diacyl- and monoacylglycerollipases into glycerol and free fatty acids, including
arachidonic acid. 2 At least ten distinct PLC isoforms have been cloned and sequenced."
The second step in prostaglandin synthesis is the oxygenation and reduction of
arachidonic acid to form an unstable intermediate, endoperoxide (PGH~. 5 This step is
catalyzed by prostaglandin endoperoxide H synthase (PGHS). Two forms of PGHS have
been identified: PGHS-1 and PGHS-2, each having cyclooxygenase and peroxidase activities.~>
The principal feature distinguishing PGHS-1 and PGHS-2 is the regulation of their e.xpression.
PGHS-1 is a gene product constitutively expressed under normal physiological conditions.
Levels of PGHS-1 increase late in gestation and peak at term, but do not rise in response to
labour/ PGHS-2 is an inducible enzyme that is expressed after stimulation of the amnion and
chorion by cytokines, growth factors or tumour promoters.7 ,II Recent studies suggest that the
increased production of prostaglandins by intrauterine tissues during labour results from an
increase in the expression of the PGHS-2 isoform. 9"11
The third enzymatic step in prostaglandin synthesis is the conversion of the
endoperoxide intermediate to one of the biologically active prostaglandins (02, E2. F za, I~ or
thromboxane A2• Individual cell types contain one primary isomerase, reductase or synthase
that converts the endoperoxide to a specific prostaglandin characteristic of that cell. An
additional factor involved in modulating levels of active prostaglandins is the regulation of the
rate at which these compounds are metabolized. Chorionic cytotrophoblasts contain abundant
levels of the prostaglandin-metabolizing enzyme prostaglandin 15-hydroxydehydrogenase
(PGDH), which catalyses oxidative inactivation of E and F series prostaglandins. 12
16
Prostaglandin Receptors
Prostaglandins exen their action through specific membrane receptors. The
regulation of uterine smooth muscle contraction is controlled by the quantity and distribution
of prostaglandin receptors as much as by the tissue concentration of biologically active
prostaglandins. The prostaglandin E receptor has four subtypes: EP 1, EP 2, EP 3 and EP _..
These couple to two major effector pathways. 13 EP 1 and EP 3 receptors promote muscle
contraction through mechanisms that include increased calcium utilization and inhibition of
intracellular cyclic adenosine monophosphate (cAMP). EP 2 and EP _. act through increased
cAMP formation to cause a relaxation of smooth muscle . H Prior to term, the action of EP 1
receptor is a factor in maintaining myometrial quiescence.
The expressiOn of myometrial prostaglandin receptors increases during late
gestation. 15 At term, EP 1 and EP 3 receptors in the uterine fundus act to promote uterine
contractions. EP 2 and EP _. receptors are concentrated in the lower uterine segment and
cervix and may facilitate cervical effacement and dilation. Misoprostol is a synthetic
prostaglandin E1 analogue that is rapidly absorbed following oral administration. It is de
esterified to the active metabolite misoprostol acid. Misoprostol acid binds to EP 3 and EP,.
receptors. l6 When used as a labour induction agent, it can effect contraction of the uterine
fundus and relaxation of the lower uterine segment and cervix.
Challis et al.17 described compartmentalization of prostaglandin synthesis and
metabolism within the human fetal membranes, decidua and myometrium in late gestation. In
the amnion, prostaglandin H 2synthase (PGHS) activity predominates, with an increase in the
e..xpression of the PGHS-2 isoform at the time of labour. Prostaglandin E2 is the principal
17
prostaglandin formed in the amnion. The chorion contains high concentrations of PGHS and
the prostaglandin-catabolizing enzyme 15-hydroxyprostaglandin dehydrogenase (PGDH). A
reduced activity of PGDH has been shown to be associated with preterm labour. 18
During labour, PGDH undergoes a reduction in activity in the portion of fetal
membranes overlying the cervical os.19 The reduced activity of this enzyme in the region of
the cervix is believed to potentially facilitate passage of active PGE1 from the amnion across
the chorion, thereby facilitating cervical ripening. Funhermore, abnormal persistence of this
enzyme might hinder the ripening process by diminishing the concentration of prostaglandins
within the cervix. 10 The maternal decidua is a source of significant prostaglandin synthesis;
however, the levels of PGHS-1 and PGHS-2 do not change during gestation or with labour
onset. It has therefore been postulated that one of the phospholipases may be the primary
rate-determining factor in decidual prostaglandin synthesis. 11
Prostaglandins and Parturition
Our knowledge of the mechanism of parturition in humans is increasing but not all
relevant pathways have been completely elucidated. Prostaglandins clearly play an imponant
role in the human labour process; however, a precise and complete description of their
mechanism continues to be sought. They appear to be active in the triggering or facilitation of
a complex sequence of events involving the fetus, fetal membranes and maternal tissues. tt-.u
The role of the human fetus in initiating labour is not well defined. In animal studies,
it has conclusively been shown that the fetal sheep triggers the onset of labour through
activation of the fetal hypothalamic-pituitary-adrenal axis (HP A), resulting in increased
18
secretion of cortisol from its adrenal gland.11 Fetal cortisol acts on the placenta to upregulate
prostaglandin synthesis, and to change the pattern of steroidogenesis, such that progesterone
levels fall and estrogen levels rise. Once activated, the ovine myometrium is able to respond
effectively to stimulation provided by the increased production of uterotonic agents such as
oxytocin and prostaglandins. While a similar mechanism has not conclusively been
demonstrated in humans, homology to the human parturition process is hypothesized.
In humans, the bioactivity of circulating corticotropin-releasing hormone (CRH) is
diminished during pregnancy by the presence of a high-affinity CRH-binding protein (CRH
BP) in maternal blood.n This blocks the action of CRH in promoting ACTH release from
pituitary cells and inhibits the stimulatory effect of CRH on prostaglandin production by
intrauterine tissues. CRH-BP levels fall near term, with a corresponding increase in the activity
of CRH. Studies in primatesn-29 have suggested that cortisol of fetal origin acts to promote
delivery. Cortisol activates amniocytes, cytotrophoblasts and decidual cells to express CRH,
further enhancing prostaglandin production by these cells. In addition, elevations in CRH may
act synergistically with oxytocin or prostaglandins to stimulate myometrial contraction. 211
Recent studies have demonstrated that exogenous glucocorticoid administration stimulates
expression of CRH by placenta, decidua and fetal membranes, which in turn enhances
prostaglandin synthesis and impairs prostaglandin metabolism in these tissues, promoting
parturition. n
19
Prostaglandins and Cervical Ripening
In 1947 Danforth30 characterized the connective tissue component of the cervi.x and
recognized that changes in its structure and biochemistry were key elements of cervical
ripening. The mechanism of cervical ripening involves a series of biochemical events distinct
from those responsible for myometrial activation and similar to those observed in tissue
inflammation.31 The trigger responsible for initiating the biochemical events that lead to
cervical ripening has not yet been identified. Prostaglandins play an important role in cervical
ripening; however, the molecular actions of prostaglandins to effect cervical ripening has not
been fully defined.
Smooth muscle and fibroblasts make up the cellular component of the human cervi."X.
The extracellular matrix is composed of substances secreted by fibroblasts, such as collagen,
glycosaminoglycan and glycoproteins. The glycosaminoglycans are comprised of repeating
units of hexosamines or uronic acid linked as polymer chains. Chondroitin sulfate and
dermatan sulfate are sulphated glycosaminoglycans while hyaluronate and chondroitin are
unsulfated glycosaminoglycans. The glycosaminoglycans attach in branched patterns to
glycoproteins to form very large molecules known as proteogylcans. 32 Elastin is another
important component of the extracellular matrix of the human cervi.x. Elastin fibers are
organized parallel to and between collagen fibers and are capable of being stretched in any
direction. During pregnancy and particularly labour the uterine cervix undergoes dramatic
changes in histological characteristics and biochemical composition.
Early in gestation, cervical hyperplasia results from turnover of both smooth muscle
cells and fibroblasts. As pregnancy advances, physiological cell death occurs, inducing
20
migration of neutrophils, macrophages and mast cells into cervical tissue.31 These cells
produce inflammatory cytokines that are responsible for stimulating the production of
metalloproteinases that cause collagen degradation at term. Nitric oxide may induce
prostaglandin production by stimulating cyclooxygenase activity . n
Ellwood et al. 3 .. described the production of prostanoids by the human cervix in
pregnancy, demonstrating that the cervix produces E.:, I!, and F prostaglandins. The
production of these compounds was furthermore shown to increase at term. Norstom et al.35
postulated that prostaglandins might exen their effects on the cervix by modulating fibroblast
activity, thereby controlling the biophysical and biochemical properties of the extracellular
matri.~. They observed that cervical fibroblasts from women treated with prostaglandins
undergo the same morphologic changes observed in women at the time of spontaneous
labour.
Murota et al. 36 demonstrated that prostaglandins are capable of stimulating the
production of hyaluronic acid by cervical fibroblasts. The resulting tissue hydration alters the
glycosaminoglycan/ proteoglycan composition in favour of cervical ripening. As pregnancy
advances closer to term, multiple factors work together in comple.~ interactions that cause
collagen dispersion and ripening of the cervix. At the onset of labour, these substantial
changes in hyaluronic acid, cytokines and collagenases, combined with the mechanical force of
uterine contractions extend cervical elastin and allow dilation of the cervbc required for
parturition.
21
Cbapter 3
MECHANICAL METHODS OF LABOUR INDUCTION
Sweeping Membranes
Stripping or sweeping of the fetal membranes was first described by James Hamilton
in 1810.37 The technique involves mechanically separating the membranes from the lower
uterus by sweeping a finger circumferentially around the interface of the fetal membranes with
the lower uterine segment at the time of pelvic e.'<amination. Sweeping of the membranes aims
to initiate labour through a cascade of physiological events, the exact mechanism of which is
unknown. McColgin et al.311 concluded that membrane stripping causes acute elevation in
plasma prostaglandin FJ.a and endocervical phospholipase AJ. activity. This release of local
prostaglandins is believed to initiate cervical ripening and the onset of active labour.
The Cochrane Library systematic review of stripping or sweeping of membranes for
inducing labour or preventing post-term pregnancy assessed the effects of membrane
sweeping to promote or induce labour on maternal and perinatal outcomes.39 Sweeping of
membranes, performed in low-risk women at term, was associated with a decreased incidence
of pregnancy continuing beyond 41 or 42 weeks' gestation (relative risk (RR) 0.42, 95%
confidence interval (CI) 0.19-0. 93) but was not an effective induction method. The reviewers
concluded that the routine use of sweeping the membranes at term did not seem to produce
clinically important benefits.
22
Studies of sweeping membranes.f0-49 suffer from heterogeneity in study design and
selection bias. Some of these studies have flaws in their design or analysis, as well as
deficiencies in reporting of results. The frequency with which sweeping was performed
varied from study to study. The first study by Swann et al . ..c described sweeping membranes
daily for three days. The authors reponed an increase in the spontaneous delivery rate from
26% to 69%. This study was not formally randomized and did not stratify based on parity
or Bishop score. lvlcColgin et al.·41 examined membrane stripping performed weekly
beginning at 38 weeks' gestation and found a significant decrease in the number of post
dates pregnancies. Patient demographics were not described, making it impossible to assess
comparability of the two groups.
The study by Weissberg and Spellacy~l suffered from selection bias with an excess
number of study group panicipants having documented Bishop scores greater than six.
Several of the studies~3.~ enrolled only post-dates patients defined as beyond 40 completed
weeks' gestation. Crane et al.~5 studied the effectiveness of sweeping membranes at 38-40
weeks' gestation based on the ability to pass a finger through the internal os and strip the
membrane by sweeping the examiner's index finger twice in a circumferential manner. No
significant difference in the median number of days to delivery was detected between the
groups (33% sweeping, 38% control, P - .39).
Concerns may arise that membrane sweeping may not have been conducted
properly in a number of these studies. As well, multiple episodes of sweeping potentially
may be more effective than a single sweep. It is possible that sweeping membranes is more
effective in post-dates than term patients or more effective in multiparous than nulliparous
patients. The heterogeneity of results in published studies comparing sweeping membranes
23
to no intervention does not allow firm conclusions to be drawn about the relative merits of
sweeping membranes for the purpose of labour induction with any degree of confidence.
Balloon Catheters
Obstetricians have used balloon catheters for more than 100 years to induce labour.
Recently, Foley catheters have been used for this purpose. 50 After thorough cleansing of the
vagina and cervix, the catheter is inserted into the endocervix and passed above the level of
the internal cervical os. The balloon is then inflated with 30-50 ml of sterile saline and left in
place for 24 hours or until spontaneous expulsion. ;r,s.:! Some authors suggest the application
of traction to the catheter or the infusion of saline through the catheter to accelerate the
ripening process.53"56
The mechanism believed to be responsible for cervical ripening is direct pressure
and overstretching of the lower uterine segment, causing local secretion of prostaglandin.
This mechanism is attested to by an increase in prostaglandin levels in maternal blood.57
Balloon catheters are absolutely contraindicated in patients with a low-lying placenta or
antepartum hemorrhage. Other relative contraindications include cervicitis and ruptured
membranes. No randomized trials assessing the Foley catheter for cervical ripening have
included patients with a history of previous cesarean delivery.
Several investigators compared the balloon catheter to intravaginal prostaglandin
gel, finding the catheter to be significantly more effective than placebo, and as effective as
prostaglandin in ripening an unfavourable cervix, as measured by the Bishop Score and
duration of labour.58..(,1 No significant difference was found in the cesarean delivery rate.
Spontaneous labour; however, occurred significantly more frequently in the prostaglandin
24
group. Atad et al.I,(J reported their experience with a special double balloon catheter and
concluded that it was as good as prostaglandin E2 {dinoprostone) vaginal tablets and better
than oxytocin in terms of change in Bishop score, interval to delivery and cervical ripening
failure rates.
St. Onge et al.61 compared 0.5 mg intracervical prostaglandin E2 gel (n- 30) with
Foley catheter (n - 36) in a small randomized clinical trial. The authors concluded that there
was no difference in efficacy between intracervical prostaglandin (n =- 28) and intracervical
Foley catheter (n - 34) as measured by the mean change in the Bishop score. The induction
to delivery interval was significantly shorter in the Foley catheter group (P - .01).
Chamberlain et al.62 compared outpatient intracervical prostaglandin E2gel (n""'
67) with intracervical Foley catheter insertion overnight (n - 62). Both the intracervical
Foley catheter and the intracervical prostaglandin E2 gelled to similar changes in Bishop
score. The Foley catheter was more commonly associated with cervical dilatation of three to
four centimetres despite the requirement for significantly more oxytocin to induce labour (P
= .003). There was no difference in route of delivery or adverse maternal events.
A few randomized trials have compared intracervical Foley catheter plus extra
amniotic saline infusion to intravaginal or intracervical prostaglandin E2 for cervical ripening
in women with an unfavourable cervix. 53-56 The results of these studies were contradictory
and inconclusive. In two studies, fewer ripening failures occurred in the Foley catheter
group. 53.5-t L yndrup et al. 56 reported that prostaglandin E2 was more efficient in inducing
regular uterine contractions in nulliparous women; furthermore, Foley balloon catheter
ripening was associated with a longer latent phase of labour. Two of these trials 53.55 found
25
no difference in cesarean delivery rates and two reported an increased cesarean delivery rate
in those patients randomized to the balloon catheter group. 54•56
The advantages of cervical ripening with extra-amniotic balloon catheters include:
simplicity, low cost, reversibility and lack of severe maternal or fetal side effects when
compared with intravaginal prostaglandin E2_ The disadvantages include pain and
discomfort, decreased maternal mobility and a greater need for oxytocin augmentation. The
data further indicates that prostaglandin use is more likely to result in spontaneous labour
and less likely to result in failed induction. The heterogeneity of results in published studies
comparing prostaglandin E2 with Foley catheter does not allow firm conclusions about their
relative merits to be drawn with any confidence.
Synthetic Hygroscopic Dilators
Hygroscopic dilators made from either natural or synthetic materials are inserted
into the cervical canal under direct visualization using an aseptic technique. The dilators are
placed in the endocervix and kept in place by sterile gauze placed in the vagina. These
devices take up fluid from the surrounding tissue, gradually swell and ultimately cause
dilation of the cervical canal. The stretching of the cervix causes local prostaglandin release.
These devices are usually contraindicated in cases of ruptured membranes, cervicitis and
vaginal bleeding. Their use in patients with a history of previous cesarean has not been
studied.
Research groups6J-M have investigated synthetic hygroscopic dilators such as
Dilapan® (Gynotech, Middlesex, NJ) and Lamicel® (Cabot Medical Group, Langhorn, P A)
to assess their effectiveness as cervical ripening agents. Dilapan® was shown to cause a
26
significant change in Bishop score within six to twelve hours of insertion. This change was
comparable to prostaglandin and better than placebo. In a large randomized clinical trial of
444 women at term, the use of prostaglandin was associated with a higher rate of successful
induction, shorter labour, less use of oxytocin and fewer infections in the mother and
newborn when compared to Dilapan®l>4.
Natural Hygroscopic Dilators
Natural laminaria are made from dried seaweed. In controlled studies, the
overnight insertion of laminaria was found to cause significant change in the Bishop score in
patients with unfavourable cervixes. 65-69 In two of these studies, patients with preinduction
ripening with laminaria had higher rates of successful induction and shorter induction-to
delivery intervals compared with controls receiving only oxytocin."5'66 These findings were
contrary to findings of two other groups. 67•
611 A recent randomized clinical trial comparing
laminaria to extra-amniotic saline infusion reported that laminaria shortened the interval to
delivery by eliminating the need for a preinduction ripening interval.67 No significant
difference was noted in rates of cesarean delivery or adverse maternal or neonatal outcomes.
Induction of labour is often indicated when the benefits to either the mother or the
fetus outweigh the benefit of continuing the pregnancy. Obstetricians routinely use various
mechanical methods to ripen the cervix; however, no single mechanical approach has
enjoyed universally superior success or acceptance. The consistency of the cervLx is dearly
related to the rate of success of labour induction and to the duration of the induction-to
delivery interval. Women with the most unfavourable cervixes still face high rates of
induction failure and cesarean delivery. The safety of mechanical methods for induction of
27
labour has not been established. There is a need for properly designed and conducted
randomized clinical trials to evaluate the efficacy and safety of many of the mechanical
methods of labour induction reviewed here.
28
Chapter 4
PHARMACOLOGIC INDUCTION AGENTS
Pharmacological Induction Agents
Pharmacological techniques for induction of labour were first described over 400
years ago. 70 One of the first recorded references described the use of juniper berries, cinnamon
and castor oil to hasten binh. In 1906, Sir Henry Dale discovered uterotonic bioactivity in
e.xtracts of the posterior pituitary gland. William Blair Bell administered a crude preparation of
these extracts to pregnant women and reponed the first clinical studies on oxytocin use for
induction of labour in 1909.71 Pierce and Du Vigneaud determined the structure of oxytocin
in 1950, for which Du Vigneaud was later awarded a Nobel Prize in chemistry. Charles Liebn
first discovered prostaglandins in 1930 but it was not until the late 1960s and early 1970s that
prostaglandins E1 and E2 were first used clinically for induction of labour.
Since the discovery of prostaglandins and their use as agents for induction oflabour,
a wide variety of different formulations, routes of administration and dosage regimens have
been investigated. Currendy, prostaglandin PGE2 (dinoprostone) is the most widely used for
the purpose of induction, with the preferred route of administration per vagina. A variety of
vehicles have been developed for the delivery of this agent, including lactic acid-based
pessaries, water-soluble gels and most recendy, a slow -release hydrogel polymer.
29
Dinoprostone
Dinoprostone (PGE.J intracervical and intravaginal gel has been the prostaglandin
agent of choice for several decades and is currently the only pharmacological agent approved
by the U.S. Food and Drug administration and the Therapeutic Products Program of Health
Canada. It is available in three forms: a 0.5 mg endocervical gel {Prepidil®; Upjohn
Pharmaceuticals, Kalamazoo, MI), a 1.0 mg or 2.0 mg vaginal gel {Prostin®; Upjohn
Pharmaceuticals, Kalamazoo, MI) and a controlled-release prostaglandin vaginal insert
available in the United States (Cervidil~; Forest Pharmaceuticals, St. Louis, MO), Canada and
Europe (Propess®, Ferring Pharmaceuticals AB, Malamo, Sweden).71•74
A meta-analysis of selected randomized trials comparing dinoprostone with placebo
or with no treatment has concluded that the use of prostaglandins for cervical ripening is
effective and reduces the induction-to-delivery interval as well as the operative vaginal delivery
rate.75 There was a higher rate of uterine hypertonus and uterine tachysystole in those patients
who received prostaglandin (PGE.J compared to oxytocin and a trend toward higher rates of
abnormal fetal heart rate patterns. There are no randomized trials evaluating the required
duration of fetal heart rate (FHR) and uterine contraction monitoring after prostaglandin gel
dosing.
Recent studies have noted a higher rate of successful induction, greater change in
Bishop score and shorter induction-to-delivery interval with the use of vaginal gel (Prostin®)
compared to intracervical gel (Prepidil®).76 There is insufficient evidence to make firm
conclusions about the relative effects of prostaglandins and oxytocin on substantive maternal
and neonatal adverse outcomes.
30
Controlled-Release Insert
The controlled-release insert consists of a polymer base containing 10 mg of
dinoprostone with a polyester retrieval system. The insert is placed in the posterior fornix of
the vagina and releases 0.3 mg per hour of prostaglandin E2 over a twelve-hour period. It is
removed with the onset of labour, at spontaneous rupture of membranes, with excessive
uterine activity or after twelve hours. Theoretical advantages include ease of administration
and the ability to remove the medication if excessive uterine activity occurs.
Although widely used, these preparations may cause excessive uterine activity leading
to placental insufficiency and abnormal fetal heart rate changes, and are associated with greater
cost and more elaborate storage requirements. Dinoprostone must be refrigerated until shortly
before administration. The average wholesale price ofPrepidil® is $48.98 per 0.5 mg dose,
Prostin® $54.00 per 2.0 mg dose and $84.00 for the 10 mg vaginal insert.
Several author?·79 have compared Cervidil® to placebo and demonstrated that it is
effective in cervical ripening but is associated with a higher incidence of excessive uterine
activity and hyperstimulation when compared to prostaglandin E2 intracervical gel. In a recent
meta-analysis performed by our research group, 80 we e."<amined the effectiveness and safety of
controlled-release prostaglandin for cervical ripening in nine randomized trials that compared
controlled-release vaginal prostaglandin (with a retrieval string) with any other cervical
ripening agent or with placebo. After assessing the homogeneity of results, the summary odds
ratio and confidence intervals were determined.
31
Compared with placebo, controlled-release prostaglandin resulted in cervical change
(summary odds ratio (OR)- 3.99, 95% (CI)- 2.71, 5.86), a higher rate of vaginal delivery in
12 hours (OR- 29.01, 95% CI- 7.08, 118.87) and less need for oxytocin (OR== 0.14, 95%
CI 0.06, 0.32) but a higher incidence of excessive uterine activity (P < .001) and
hyperstimulation (P = .004). When compared with Prepidil®, there was a higher rate of
excessive uterine activity with controlled-release prostaglandin (P == .03), but less need for
oxytocin (OR - 0.09, 95% CI - 0.01, 0.53). With Prepidil® plus immediate oxytocin, there
was a lower rate of active labour in twelve hours compared with controlled-release
prostaglandin (OR - 0.27, 95% Cl== 0.10,0.72).
There was a lower rate of vaginal delivery in 12 hours and a higher incidence of
oxytocin use with controlled-release prostaglandin as compared with misoprostol (OR = 0.53,
95% CI - 0.34, 0.83 and OR - 1.58, 95% CI 1.08, 2.32, respectively). The induction-to
delivery interval was shorter with controlled-release prostaglandin than with placebo or
Prepidil® but longer than with Prepidil® plus augmentation with immediate oxytocin or
misoprostol. Although no differences in maternal or fetal morbidity were noted, the sample
size was not adequate to evaluate these outcomes. Controlled-release prostaglandin is an
effective cervical ripening agent as compared with Prepidil® but may result in an increased
incidence of excessive uterine activity. Controlled-release prostaglandin may not be as
effective as misoprostol or Prepidil® plus immediate oxytocin. The American College of
Obstetricians and Gynecologists recommends that the fetal heart rate and uterine activity be
continuously monitored electronically for the duration of insert placement.11t
32
Misoprostol
Misoprostol (Cytotec®; Searle, Oakville, ON), a synthetic prostaglandin (PGE J
analogue, is marketed in an oral formulation of 100 ~-&g or 200 ~-&g tablets.82 It is currently
approved by the U.S. Food and Drug Administration for use in the treatment and prevention
of gastrointestinal ulcer disease resulting from non-steroidal anti-inflammatory use.
Misoprostol is an inexpensive prostaglandin capable of initiating uterine contractions;
therefore, it has been extensively investigated for use as an induction agent in pregnancy. A
single 100 J,lg tablet costs $ 0.30; therefore, a single 50 ~g dose may cost as little as $0.15.
Misoprostol is also temperature-stable and does not have special storage requirements.
Vaginal misoprostol is currently widely used in the United States for cervical ripening and
labour induction; however, investigations continue regarding the optimal dose, dosing regimen
and route of administration.
Review of Vaginal Misoprostol Literature
Since the first clinical trial of vaginal misoprostol for labour induction reported by
Margulies et al. in a letter to the Lancet in 1992, there has been growing interest in misoprostol
for use as a labour induction agent.113 In this randomized trial, 64 women beyond 28 weeks'
gestation undergoing induction of labour were given 50 ~g doses of intravaginal misoprostol
or intravenous oxytocin. Successful induction was defined as vaginal delivery within 24 hours
of the start of induction. More women in the misoprostol group achieved successful vaginal
delivery within 24 hours than in the oxytocin group (79% compared with 62%). The mean
time (± standard deviation) to vaginal birth delivery 407 (± 265) minutes in the misoprostol
33
group and 577 (± 605) minutes in the oxytocin group P = 0.02. There were no differences in
mean birth weight or Apgar scores. No adverse fetal or neonatal outcomes were reported;
however, this small study did not have sufficient power to form conclusions about maternal or
fetal risks. The authors concluded that intravaginal misoprostol was more effective than
oxytocin for labour induction in the third trimester of pregnancy.
Several randomized trials 8>97 comparing vaginal misoprostol to standard therapy
(dinoprostone) found misoprostol to be safe and effective for induction of labour. In 1993
Sanchez-Ramos and colleagues8 .. compared 50 J.tg of intravaginal misoprostol administered
every four hours to intravenous oxytocin in 129 women with uneffaced and undilated cervixes.
The interval from the start of induction to delivery was significantly shorter in the misoprostol
-treated group (11 versus 18 hours, P == .004); however, the frequency of uterine tachysystole
(defined as six contractions in a 10 minute period, repeated over two consecutive 10 minute
periods) in the misoprostol group was approximately three times that found in the oxytocin
treatment group (34.4% versus 13.8%, P < .05). No differences were found in delivery route,
neonatal or maternal morbidity.
The authors concluded that vaginal misoprostol was safe and effective for labour
induction and recommended further investigation to detail the optimal route, dose and dosing
regimen. The frequency of excessive uterine activity was evaluated as a secondary outcome in
the study, and did not form a basis for calculation of sample size. Investigators who reviewed
fetal tracings were not blinded to group assignment. The non-blinded study design and the
subjective nature of the evaluation and classification fetal heart rate tracings introduces bias
34
and raises concerns about inter-observer and intra-observer variation. This study did not have
sufficient power to draw firm conclusions about the safety of misoprostol.
In 1995, Winget al. 85 compared the administration of intravaginal misoprostol50 JJg
every three hours with intracervical dinoprostone 0.5 mg every six hours in 135 women with
unfavourable cervixes (Bishop score less than 4). Misoprostol was found to be more effective
for labour induction. The mean time (±standard deviation) to vaginal birth was 903 (± 482)
minutes in the misoprostol-treated subjects and 1410 (± 482) minutes in the control group (P
< .001). The misoprostol-treated subjects required oxytocin augmentation much less
frequently than those treated with dinoprostone 33.8% versus 65.7%, (P < .001). More than
70% of women in the vaginal misoprostol group delivered vaginally within 24 hours compared
with 47% of dinoprostone-treated women (P < .01).
Misoprostol was associated with a 36% incidence of tachysystole (defmed as greater
than six or more uterine contractions occurring in 10 minute window for two consecutive 10-
minute periods) compared with a 12% incidence of tachysystole in the dinoprostone treated
group (P < .001). A 30% incidence of passage of meconium in utero was noted in the
misoprostol group compared with a 10% passage of meconium in the dinoprostone-treated
group (P < .05). No significant difference in maternal or neonatal outcomes was found
between the two groups.
The limitations of this study include, lack of blinding, an empiric dosing regimen and
lack of rigor in defining the treatment of study group and controls. The 50 f.lg dose of
35
misoprostol every three hours was extrapolated from a previously published study of
misoprostol. 8.. There was a difference in the time intervals from last dose of medication to
allowable administration of oxytocin which could have potentially biased the primary outcome.
Oxytocin administration was permitted three hours after the last dose of misoprostol and six
hours after the last dose of dinoprostone. The frequency of excessive uterine activity and
meconium staining were evaluated as secondary outcomes in this study. Investigators who
reviewed fetal tracings were not blinded to group assignment. The non-blinded study design
and the subjective nature of the evaluation and classification fetal heart rate tracings introduces
bias and raises concerns about inter-observer and intra-observer variation. This study did not
have sufficient power to draw firm conclusions about the safety of misoprostol.
In 1996, Mundie et al.116 compared 50 !Jg vaginal misoprostol administered every
four hours to standard labour induction methods (dinoprostone and or oxytocin) in 222low
risk women undergoing cervical ripening and labour induction at term. The mean time to
vaginal delivery (±standard deviation) was 753 (± 588) minutes for vaginal misoprostol versus
941 (± 506) minutes for intracervical or intravaginal dinoprostone. Oxytocin augmentation
was used less frequendy in the misoprostol group (RR = 0.48, 95% CI 0.31, 0.7 4). There were
no significant differences in cesarean rate. Maternal and neonatal outcomes were not
significandy different.
The major weaknesses of this study were lack of blinding and rigor in defining the
standard protocol. The control group could receive one of several options: Prepidil® 0.5 mg
intracervically every six hours, Prostin® 1 or 2 mg intravaginally every six hours or a dilute
36
solution of oxytocin administered intravenously. The choice of the agent to be used in the
control group was to be chosen by the individual attending physician caring for that patient
based on whichever option the physician felt was optimal for the care of that particular patient.
This study did not have sufficient power to draw firm conclusions about the safety of
misoprostol.
In a subsequent clinical trial involving 276 women, Winget al. 87 used a lower dose of
misoprostol, comparing 25J,lg of intravaginal misoprostol every three hours with 0.5 mg of
dinoprostone administered intracervically every six hours in an attempt to reduce the
occurrence of tachysystole and meconium passage noted in the misoprostol group in their
earlier trial.85 Misoprostol was found to be effective for labour induction in women with
unfavourable cervi.xes (Bishop score less than 4). The mean time (± standard deviation) to
vaginal birth was 1323 (± 844) minutes in the misoprostol-treated subjects and 1532 (± 706)
minutes in the control group (P < .05). The misoprostol-treated subjects required oxytocin
augmentation much less frequently than those treated with dinoprostone 46% versus 73%, (P
< .001).
The dosing regimen of misoprostol25 !Jg every three hours was associated with a
17% incidence of tachysystole (defined as greater than six or more uterine contractions
occurring in 10 minute window for two consecutive 10-minute periods) compared with 14%
incidence of tachysystole in the dinoprostone treated group (P ==0.08). A 17% incidence of
passage of meconium in utero was noted in the misoprostol group compared with a 14%
37
passage of meconium in the dinoprostone-treated group (P > .05). No significant difference in
maternal or neonatal outcomes was found between the two groups.
Limitations of this study include lack of blinding and a discrepancy in treatment
between the study and control groups. Allowable dosing intervals from last dose to start of
oxytocin differed in the investigational protocol, 6 hours for dinoprostone and three hours for
misoprostol. Pharmacokinetic studies would have strengthened this study considerably. The
primary outcome was the achievement of successful induction. All other outcomes were
secondary. This study did not have sufficient power to draw firm conclusions about the safety
of misoprostol.
Sanchez-Ramos et al. ~8 studied the safety and efficacy of vaginal misoprostol for
cervical ripening and labour induction in a meta-analysis of trials published up to 1997. Of 16
trials identified, eight 8"· ~s. 87
' 89-?J met their criteria and included 966 patients, 488 of whom
received vaginal misoprostol for labour induction. Controls received oxytocin or vaginal
dinoprostone. Those who received vaginal misoprostol had a higher incidence of vaginal
delivery within 24 hours of application (OR 2.64, 95% Cl 1.87, 3.71) and a lower cesarean rate
(OR 0.67, 95% CI 0.48,0.93). Use of vaginal misoprostol was associated with a higher
incidence of tachysystole (OR 2.70, 95% CI 1.80, 4.04) but not ofhyperstimulation (OR 1.91,
95% CI 0.98, 3.73).
The number of subjects allocated to the misoprostol group ranged from 24-138 with
control groups of similar size. Five of the eight trials included were conducted in the United
StatesS-t.ss.B7. 92.93 while the other three were conducted in South America.ll9-91 The proportion of
38
nulliparous patients were similar and all patients enrolled in the control group received
dinopro~one or oxytocin with the exception of one small trial where control patients received
placebo.
The primary problem with this meta-analysis was the heterogeneity in study design.
The dose of misoprostol varied from 25 J.tg every two hoursn to lOO J.Lg119·'}
0 as a single dose.
There was no one standard treatment for the control groups. In one study, placebo was used
in the control group. 89 In two studies, s ... 91 dinoprostone 0.5 mg intracervically could be used
with or without oxytocin. In the remaining four studies controls received dinoprostone 0.5
mg intracervically. 85' 87
' n. 93 In one study the dose of intravaginal dinoprostone was 3 mg.'JC
Continuous electronic fetal monitoring was performed on all patients in si-x studies
while the other two studies used fetal monitoring intermittently.119'90 None of the individual
trials had sufficient power to detect a reduction in cesarean delivery rate; however, a reduction
in overall cesarean rate was observed when the studies were considered together. The
reduction in cesarean rate in this meta-analysis depended heavily on one study.91 When the
analysis was repeated after e.xcluding each trial, it appeared that the Campos study was an
important contributor to the overall heterogeneity due to its large discrepancy in cesarean rates
between the misoprostol (3.8%) and control groups (25.3%).
In an attempt to define the optimal dosing regimen of vaginal misoprostol, Winget
al.9• compared 25 J.Lg of intravaginal misoprostol every three hours with 25 J.Lg of intravaginal
misoprostol administered every six hours in 522 women with unfavourable cervi-xes (Bishop
score < 4). In this trial, the six-hour regimen was found to be less efficacious than the three-
39
hour regimen. The mean time (±standard deviation) to vaginal binh was 903 (± 482) minutes
in the three-hour dosing group and 1410 (± 869) minutes in the si."<-hour dosing group (P <
.001). The overall frequencies of uterine contraction abnormalities were 14.6% with the three
hour dosing interval and 11.2% with the six-hour dosing interval (P < .05). A low incidence of
uterine hypenonus and hyperstimulation was observed with both regimens (4.2% with the
three-hour dosing and 4.6 with the six-hour dosing regime). The frequency of meconium
passage was lower with the longer dosing interval than that observed in earlier studies (10%).
No adverse fetal outcomes were reported; however, there was one maternal death from
amniotic fluid embolus and two cesarean hysterectomies for uterine atony. This study did not
have sufficient power to draw conclusions about maternal or fetal safety.
Sanchez-Ramos et al.95 compared 50 f.lg of vaginal misoprostol with the slow-release
dinoprostone vaginal insert. The vaginal insert was administered as a single application for a
maximum of 12 hours. Misoprostol-treated women delivered in a shorter time interval than
the dinoprostone treated women (P < .001) but e."<perienced more tachysystole (21.3% versus
7.0%, P =- .004). No differences in route of delivery, intrapartum complications or adverse
neonatal outcomes occurred between the two groups. The authors concluded that both agents
were effective and safe but that misoprostol appeared more effective and offered substantial
cost sav10gs.
Winget al.96 compared the efficacy of 25 f.lg of misoprostol administered every four
hours to a 24-hour exposure of the dinoprostone vaginal insen. Two hundred women were
enrolled. There were no significant differences in the interval from stan of induction to
40
delivery. There were no significant differences in delivery route or use of oxytocin between
the two groups. Approximately 50% of women in both treatment arms delivered vaginally
within 24 hours. Uterine tachysystole and hyperstimulation occurred more often in the
dinoprostone-treated subjects (18.4% versus 7.1%, P = .02).
The Cochrane Pregnancy and Childbirth Group'~7 reviewed twenty-six selected
randomized trials of vaginal misoprostol. They studied the effectiveness and safety of
vaginal misoprostol compared with placebo, oxytocin or prostaglandin E1 for cervical
ripening or induction of labour at term. The primary outcomes were the rate of vaginal
delivery within 24 hours, the incidence of uterine hyperstimulation with associated changes
in the fetal heart rate, the rate of cesarean delivery and the incidence of serious adverse
effects in the fetus or the mother.
When compared to dinoprostone, failure to achieve vaginal delivery within 24
hours was reduced in four of five trials (RR 0.70, 95% CI 1.17-1. 99). The reported incidence
of uterine hyperstimulation was increased with misoprostol (RR 1.59, 95% CI 1.02-2.48) as
was meconium staining of the amniotic fluid (RR 1.38, 95% CI 1.06-1.79). Rates of vaginal
instrumental delivery and cesarean delivery were inconsistent between trials. Overall, there
was a reduction of instrumental deliveries with misoprostol (RR 0.75, 95% Cl .58-.97).
There were no significant differences in perinatal or maternal outcomes.
When comparing low dose regimens of misoprostol to higher doses regimens (25
J.lg every six hours compared to every three hours or 25J.lg compared to 50J.lg every three
hours) neither group showed significantly more failures to achieve delivery within 24 hours
41
(RR 1.08, 95% CI .93-1.25). There was significantly more use of oxytocin for labour
augmentation in the low dose group (RR 1.32, 95% CI 1.11-1.56). There was a trend toward
less uterine hyperstimulation, fewer low Apgar scores at five minutes and fewer admissions
to neonatal intensive care units. There were no differences in mode of delivery, meconium
stained amniotic fluid or maternal side effects.
The Cochrane collaborative group concluded that in dosages of 25 J.Lg every three
hours or greater, misoprostol is more effective than dinoprostone and oxytocin for the
induction of labour at term. No differences in perinatal outcomes were shown; however, the
studies were not sufficiently large to exclude the possibility of uncommon serious adverse
effects. Although vaginal misoprostol showed promise as a highly effective, inexpensive
and convenient agent for labour induction, it could not be recommended for routine use at
this time.9M
9
42
Table 1: Summary of Randomized Trials of Vaginal Misop_rostol for Labour Induction
Author Year No. Misoprostol Control group Results Pts. Dose
Indue-delivery time Margulies83 1992 64 50 J,Lg q2h Oxytocin shorter (miso.)
(P- .02)
Sanchez8 ..
Oxytocin Indue-delivery time
1993 129 50 J,Lg q4h shorter (miso.) +PGE2 (P == .004)
Indue-delivery time Fletchers9 1993 45 100 J.lg Placebo shorter (miso.)
(P < .001)
Fletcher90 1994 63 100 J,Lg PGE2 (3 mg) Indue-delivery time not
different (P = n.s.) Indue-delivery time
Wing 115 1995 135 50 J,Lg q3h Prepidil ® shorter (miso.) (P < .001)
Indue-delivery time Wing9-4 1995 276 25 J,Lg q3h Prepidil ® shorter (miso.)
(P < .001) Indue-delivery time
Varaklis92 1995 68 25 J.lg q2h Prepidil ® shorter (miso.) (P .... 006)_
Indue-delivery time Chuck93 1995 103 50 J,Lg q4h Prepidil ® shorter (miso.)
(P < .001) Indue-delivery time
Mundle86 1996 222 50 J,Lg q4h PGE2 shorter (miso.) (P- .04)
Indue-delivery time Wing87 1996 522 25 J,Lg q3h 25J,Lg q6h shoner (miso. q3h)
(P < .001)
Wing96 1997 200 25 J,Lg q4h Cervidil® Indue-delivery time not different (P - n.s.) Indue-delivery time
Sanchez95 1998 223 50 J,Lg q3h Cervidil® shorter (miso.) (P < .001)
43
Review of Oral Misoprostol Literature
Only a few randomized trials assessing oral misoprostol for induction oflabour at
term have been published.1c:o-107 Ngai et al.100 compared 200 J.lg oral misoprostol to 50 mg oral
vitamin B12placebo for cervical ripening in 80 women with prelabour rupture of membranes at
term. Twelve hours after receiving the assigned treatment, oxytocin was administered as
necessary. There was a significant difference in Bishop score 5.7 (± 1.8 SD) in the misoprostol
group and 6.1 (± 1.9 SO) in the placebo group P< .05. The sample size was calculated to
detect a difference in Bishop score 12 hours after administration of medication. The
assumptions were a of .05 and 1-(3 = 95. The sample size was initially calculated to be 100;
however, an interim analysis was conducted after 80 patients were enrolled and the difference
in Bishop score was found to be highly significant in both nulliparous and multiparous
participants.
The strength of this study is its design as a double blind placebo controlled trial. The
weaknesses include the choice of primary outcome. More clinically imponant and objective
outcomes such as time to delivery would have been more relevant in this population of
women with prelabour-ruptured membranes. The authors do not explain their decision to do
an interim analysis after eighty patients were enrolled. An additional twenty patients would
have met their sample size requirement. All patients in the study received oxytocin twelve
hours after enrolment. This is a more aggressive approach than the current standard of care,
which involves administration of oxytocin within 24 hours of membrane rupture.
Windrim et al. 101 studied the effectiveness, safety and gastrointestinal tolerance of
misoprostol used orally for induction of labour and compared its use to established induction
protocols. The primary outcome measure was time to delivery. Two hundred seventy-five
women who presented for induction of labour were randomly assigned to receive either 50 #'g
of misoprostol orally every four hours or treatment with physician-chosen combinations of
either intracervical or vaginal prostaglandins every 4 to 6 hours or artificial rupture of
membranes followed by oxytocin infusion. The mean time (±standard deviation) to vaginal
birth was 926 ± (521) minutes compared to 909 ± (585) minutes with the established protocol
P =- 0.81, a non-significant difference. There were no clinically or statistically significant
differences in maternal secondary outcome measures (cesarean rate, frequency of epidural use,
perineal trauma or manual removal of the placenta). There was no difference in maternal
gastrointestinal side effects. Neonatal outcomes including cord blood acid-base analysis were
not different. This study identified oral misoprostol as a new option for labour induction and
recommended further studies to confirm the safety and efficacy and to determine optimal dose
and frequency of administration.
The sample size was calculated using a two-tailed a of 0.05 and a 1-P of95. This is a
smaller type two error than the 13 0.20 usually chosen in similar trials. These authors chose this
J3 of 5 to reduce the chance of missing a true difference between the two groups if a difference
actually did exist. The major clinical consequence of conducting a trial with insufficient power
to detect a difference in the time to delivery interval might be that patients could be offered a
new induction agent that performed less well than the standard induction agent dinoprostone.
The primary outcome was time to delivery, which is less substantive than more important
clinical outcomes such as cesarean delivery rates and fetal asphyxia. To test such outcomes
using misoprostol would require large sample sizes. A study of this nature would need to be
multi-centred and would be prohibitively e.~pensive.
45
The major weaknesses of this study were lack of blinding and lack of rigor in defining
the standard protocol. The control group could receive one of several options: Prepidil® 0.5
mgs intracervically every six hours, Prostin® 1 or 2 mgs intravaginally every six hours or a
dilute solution of oxytocin administered intravenously. The choice of the agent to be used in
the control group was to be chosen by the individual attending physician caring for that patient
based on whichever option the physician felt was optimal for the care of that particular patient.
Toppozada et al. 102 compared vaginal and oral misoprostol for induction of labour in
a small-randomized trial. Twenty patients assigned to receive lOOf.lg of vaginal misoprostol
were compared to twenty assigned to lOOf.lg oral misoprostol. When no response was noted
three hours following the first lOOf.lg dose, 200J,lg was then administered every 3 hours until a
ma.ximum dose of 1000f,1g was reached. Doubling the dose via the vaginal route was only
needed in one instance. One subject in the vaginal group and three in the oral group had
cesarean delivery because of failed induction. The authors concluded that vaginal misoprostol
resulted in a shorter time to delivery (P < .005). However, more abnormal fetal heart rate
patterns and uterine hyperstimulation occurred (P < .05). The validity of the results of this
study is limited by its small sample size and the heterogeneity in misoprostol dosing regimens.
The study does not have sufficient power to draw conclusions about maternal or fetal safety.
46
Table 2: Summary of Randomized Trials of Oral Misoprostol for Labour Induction
No. Misoprostol Misoprostol Comparison Author Year Pts. Vaginal Oral Dose Group
Dose
NgailOO 1996 80 200 ~gonce Placebo
PGE2 gel Windrim101 1997 275 50 ~gq4h q6h
T oppozada102 1997 40 100 ~gq3h 100 ~g q3h
Bennett1CJ 1998 206 50 ~gq4h 50 ~g q4h
Adair10 .. 1998 178 50 ~g q6h 200 ~gq6h
Wingtcs 1999 220 25 ~g q4h 50 ~gq4h
Butttc6 1999 108 50 ~gq4h Oxytocin
Results
Change in Bishop score greater
(miso.) (P < .05)
Time to delivery not different
(P = .81) Time to delivery
longer in oral (P < .005)
Time to delivery longer (oral) (P = .004)
H yperstimulation lower in oral
(P = .04) Time to delivery
not different (P = .12)
Tachysystole greater (oral)
{P < .01) Time to delivery
longer (oral) (P - .005}
Tachysystole not different (P - .39)
Time to delivery longer (miso.)
cP- .007)
47
Chapter 5
RESEARCH QUESTION AND METHODS
Research Question
The primary research question in this study was whether there existed a four-hour
difference in the interval to vaginal birth with oral misoprostol (50 J.Lg} compared with vaginal
misoprostol (50 ~g) administered every four hours for induction of labour in women at term
with intact membranes. As a secondary outcome, we also studied whether the route of
administration influences the frequency of excessive uterine activity resulting in abnormal fetal
heart rate tracings. Other secondary outcomes included neonatal morbidity (as determined by
cord blood acid-base disturbances and American College of Obstetricians and Gynecologists
criteria for binh asphyxia.), cesarean binh, maternal gastrointestinal side effects and patient
satisfaction.
Scientific Rationale and Originality
Prostaglandins are known to be effective induction agents when administered
vaginally. A meta-analysis of randomized clinical trials found vaginal misoprostol to be safe
and effective in cervical ripening and labour induction in women at term with intact
membranes.88 No masked randomized clinical trials comparing oral misoprostolSO J.Lg every
four hours to vaginal misoprostol50 ~g every four hours were found in a Medline search of
the English language literature.
48
Research Design
This study was a masked randomized placebo-controlled trial.
Sample Size Calculation
For the primary outcome, time to vaginal binh, a sample size of 172 was calculated
using a two-tailed a = .05, f3 = .20, .!\ = 240 minutes, and cr derived from pooling of results
from a cohon of women who received vaginal misoprostol 86 (588 minutes) and the oral
misoprostol intact membrane stratum 1c1 (521 minutes) in previous studies by our group
(PEPI, Version 2,1995; Computer Programs for Epidemiologic Analysis, Stone Mountain,
Georgia). Since our previous studies had consistent cesarean rates of 12.2% with 99%
confidence interval (CI) {8.8, 16.5%), we added 20% {34 patients) to allow for anticipated
cesarean binhs in our sample. A sample size of 206 was deemed appropriate to detect the
clinically imponant difference {240 minutes), determined by a survey questionnaire of patients
and medical personnel.
Setting
The study was conducted at the Grace General Hospital, St.John's, Newfoundland,
Canada, from March 3, 1997 to October 8, 1997. This referral centre for perinatal care serves
an almost entirely white population of 500,000 and is the site of 2500 of the province's 6000
annual births. Our induction rate is just over 20%, with approximately 520 inductions per
year.
49
Inclusion and Exclusion Criteria
Patient inclusion criteria were: an indication for induction, single live fetus, gestation
longer than 37 completed weeks, intact membranes and cephalic presentation. Patients who
presented with any one of the following were ineligible: non-reassuring fetal bean rate tracing
(FHR), previous uterine surgery, known hypersensitivity to prostaglandins, age less than 18
years, contraindication to vaginal birth or unwilling to consent.
Recruitment
Before staning the trial, the research protocol and study design was presented to
obstetricians, residents and labour and delivery nurses at the annual Obstetrical Resident
Research Day. The research nurse provided in-service education to the labour and delivery
staff and prenatal educators who were involved in the study. All pregnant women who
attended prenatal classes were informed of the study and given written information about the
protocol. Eligible patients were informed when induction of labour was indicated. At
presentation for induction, patients were enrolled in the study by the attending physician, who
explained the protocol and potential risks and benefits before obtaining written consent.
Randomization
Opaque envelopes containing a card indicating group allocation were prepared by an
administrative staff member using computer-generated random number tables with
randomization in blocks of four. Stratification was based on parity and Bishop score.
Allocation of each consenting patient to the induction method was determined at the time of
induction by opening the next sequentially numbered envelope. A pharmacy technician
50
prepared study medications and placebos. For each patient, eight powdered SO J.lg doses of
misoprostol were placed individually in paper packets and stored in an airtight plastic
container. Eight doses of an indistinguishable powdered cellulose placebo were similarly
prepared and separately packaged for each patient. Oral or vaginal use was specified on the
label.
Allocation Concealment
To conceal appearance and taste, all oral powders were mixed with 30 ml of apple or
orange juice according to patient preference. Vaginal powders were suspended in
hydroxyethyl gel just before use and delivered to the posterior fom.L"< using a vaginal applicator.
When a subject required more than eight doses, another identically prepared medication and
placebo package was provided. To maintain allocation concealment, research personnel who
prepared the envelopes and medications were not involved in any other aspect of the study.
The induction method was concealed from all caregivers, participants and investigators until
data analysis was completed. The code for group assignment was not broken until data
analysis was complete.
Intervention
After enrolment, subjects were e."<amined to determine Bishop score and stratified to
either the low Oess than 7) or high (7 or more) Bishop score group, and according to
nulliparity or multiparity. Each patient was then randomly assigned to receive either oral
misoprostol (50 J.l&) and vaginal placebo or vaginal misoprostol (50 J.Lg) and oral placebo every
51
four hours until the occurrence of one of the following: a contraction frequency of three per
10 minutes, a non-reassuring FHR tracing, spontaneous rupture of membranes or delivery. All
study inductions were conducted immediately after randomization on an inpatient basis, to
allow continuous electronic FHR and uterine contraction monitoring for at least one hour
after each dose. Decisions regarding amniotomy, analgesia, epidural anesthesia and oxytocin
augmentation were made by the attending physician.
Compliance
All participants enrolled in the study were accounted for, with none lost to follow-up.
After each delivery, the study package was returned to the pharmacy technician to ensure each
participant had received the assigned treatment.
Primary Outcome
Substantive maternal and neonatal primary outcomes, such as rates of cesarean birth
and binh asphyxia, are relevant when evaluating any induction agent. To test such outcomes
using misoprostol would require sample sizes of 3400 and 5300, respectively because of the
infrequency of these events. A study of this nature would need to be multi-centred and would
be prohibitively expensive. For this study, the primary outcome measure was chosen to
address clinical effectiveness. The primary objective was to compare the efficacy of
misoprostol administered orally to misoprostol administered vaginally in patients requiring
cervical ripening for the induction of labour at term. The purpose was to determine if route of
administration impacts upon the induction-to-delivery interval.
52
Secondary Outcomes
To determine whether hypertonus, tachysystole or hyperstimulation were asS<..'Ciated
with route of administration of misoprostol, all FHR graphs were reviewed before study
induction groups were unmasked. Each of the physician authors {two maternal fetal medicine
attending physicians and two senior obstetrical residents) independently classified monitoring
strips using terminology defmed by Curtis et al.1::1 as follows: hypertonus, a contraction lasting
more than 90 seconds; tachysystole, a contraction frequency of more than five in a ten-minute
period; hyperstimulation, exaggerated uterine response with late FHR decelerations or
tachycardia greater than 160 beats per minute. Planned treatment of tachysystole and
hyperstimulation included change in maternal position to the left lateral decubitus position,
oxygen administration by nasal prongs and intravenous ritodrine as needed. Fetal scalp blood
sampling was performed when indicated by non-reassuring FHR graphs, at the discretion of
the attending physician.
To assess fetal safety issues, cord blood samples for arterial and venous acid-base
analysis were taken after each delivery. Neonatal outcomes such as Apgar score at one and
five minutes were recorded. All neonates were evaluated to assess if they met the American
College of Obstetricians and Gynecologists (ACOG) criteria for birth asphyxia. 1cs To assess
patient satisfaction, all patients enrolled in the study were asked to complete a modified labour
agentry scale questionnaire. tC<J
53
Data Analysis
Data were analyzed on an intent-to-treat basis by parametric and nonparametric
statistics, using Statistix 4.1 (Analytic Software, Tallahassee, Florida). Decision levels and
hypotheses to be tested were preset to minimize bias. Hypothesis testing was performed on
the primary outcome. Other comparisons were considered hypothesis-generating. Statistical
significance of the primary outcome measure was assessed using Student's t-test and was
considered significant at P < .05. Rank order nonparametric statistics using the Mann-Whitney
U test, where cesarean birth is a failure to deliver vaginally and ranked longer than any vaginal
birth, allowed inclusion of all births in a secondary analysis with median time to vaginal birth as
the measure of central tendency.
Secondary outcomes were analyzed by parametric and non parametric statistics as
appropriate. Descriptive statistics were used for demographic baseline data. The significance
level for all secondary and hypothesis-generating analyses was set at P < .001 to account for
multiple testing, a conservative approach.
Confidentiality and Ethical Issues
Specific steps were taken to protect the rights of participants in the clinical trial. All
pregnant women who attended prenatal classes were given written information early in their
pregnancy. All participants who presented at the time of induction were given written
information sheets explaining their participation in the clinical trial. The physician responsible
for initiating the induction obtained written informed consent. Members of the research team
reviewed records concerning labour, delivery and hospital stay. No records bearing the name
54
of any research panicipant was provided to anyone other than the research team involved in
the study. No panicipant was identified in publication.
The risks associated with the study were carefully explained at the time of obtaining
written consent. All patients were informed of the potential risk of excessive uterine activity
associated with the use of prostaglandins. The planned medical treatment available and the
possibility of cesarean delivery were explained in detail. Participants were also informed of the
potential gastrointestinal side effects associated with administration of oral prostaglandins.
The Human Investigation Committee of the Faculty of Medicine, Memorial University of
Newfoundland, and the Grace General Hospital approved the research proposal.
Study Protocol for Misoprostol Induction
The study protocol for this randomized controlled trial was posted in the labour and
delivery area and served to guide medical and nursing personnel involved in recruitment of
patients and administration of study medications.
Misoprostol (Cytotec®) is a prostaglandin E1 methyl analogue with potent cervical
ripening and uterotonic effects. Randomized clinical trials in our centre have found
misoprostol to be more effective than standard induction regimes. Patients being delivered by
a family physician require consultation with an obstetrician before recruitment.
Contraindications
• Major uterine surgery
• Cephalopelvic disproportion
55
• Non-reassuring fetal heart rate tracing
• Fetal malpresentation
• Known hypersensitivity to misoprostol or prostaglandins
• Any contraindication to vaginal delivery
• Age less than 18 years.
Note
The use of prostaglandins for cervical ripening or induction in a patient with a history
of a previous cesarean delivery has not been approved by the Society of Obstetricians and
Gynecologists of Canada (SOGC) or by the American College of Obstetricians and
Gynecologists. (ACOG).
Procedure
1. The patient should be fully informed of the indication for induction.
2. The proposed use of misoprostol should be explained.
3. Written informed consent should be obtained.
4. All misoprostol inductions will be done as in-patients.
5. A written order is necessary for the initiation of the induction by misoprostol.
6. A specific order is necessary for each additional dosage of misoprostol, after physician
evaluation of the appropriateness of this dosage.
7. Vital signs are to be checked prior to administration of medication.
8. Fetal heart rate tracing (FHR) prior to misoprostol should be obtained.
9. All patients should have a Bishop Score determined and recorded prior to the start of
the induction.
56
10. The data sheet should also be stamped with the patient's admission card.
11. Fetal heart rate tracings should be done continuously for two hours after each dose of
misoprostol.
12. Repeat misoprostol dosage may be continued until one of the following occurs: labour
is established, a contraction frequency of 3 per 10 minutes.
13. Amniotomy (ARM) can be performed in either group at the discretion of the attending
physician.
14. Following ARM, misoprostol can no longer be used.
15. Augmentation with oxytocin is permitted following the use of misoprostol.
16. Oxytocin should not be started within four hours of misoprostol administration.
17. Misoprostol may be ordered in a dose of 50 micrograms up to every four hours as
needed.
18. All newborns should have cord blood gases obtained (venous and arterial).
19. The misoprostol induction data sheet should be completed for ongoing monitoring
and quality assurance of our misoprostol protocol.
Budget and Grant Applications
In an attempt to procure funding for this study, our research group submitted a
proposal to the Medical Research Council of Canada (see appendi..x F). Unfortunately, this
application was unsuccessful. A subsequent research proposal submitted for the Memorial
University and Health Care Corporation research award was successful and resulted in a
57
research grant of $25,000. The Health Care Corporation hired a departmental part-time
research nurse to assist with the study. The Budget is attached as appendix G.
58
Chapter 6
RESULTS
Study Population
During the study period, 393 patients presented for induction of labour, of whom
308 were eligible. Seventeen patients refused enrolment because they did not want to be
involved in a research protocol and the remaining 85 were not invited to participate because
alternate methods of induction including amniotomy, oxytocin or dinoprostone were preferred
by their attending physicians. Among 206 women enrolled, there were 104 in the oral group
and 102 in the vaginal group (Figure 1). No panicipant or neonate was lost to follow-up.
59
Figure 1: Study population
308 Eligible
17 -Patient refused
85 - Not invited to participate
206 - Enrolled I
I I
104 102 Oral misoprostol Vaginal misoprostol
Primary Outcome
The mean time (±standard deviation) to vaginal binh with oral misoprostol was 1072
(± 593) minutes compared with 846 (± 385) minutes with the vaginal protocol, a statistically
significant difference (P - .004). The median time to vaginal binh in the oral misoprostol
group was 1125 minutes compared with 926 minutes in the vaginal misoprostol group (P =
.38, Mann Whitney U test). When given orally, a median of two and a ma.ximum of nine doses
were used. Only seven subjects took more than five oral doses. When misoprostol was placed
vaginally, a median of two and a maximum of five doses were used. Maternal preinduction and
60
neonatal demographic data are given in Table 3. There were no significant differences in
demographic characteristics between the two groups.
Table 3: Demographic characteristics
Vaginal Misoprostol Oral Misoprostol n =102 n = 104
Nulliparous 74 69 Maternal age (years) 28.7 4.9) 27.5 5.0) Gestation (days) 284.2 (8.3) 285.4 (7.6) Gravidity 1.6• 0.8 1.7~ 1.0 Parity 0.3 I 0.61 0.5 I 0.8 Bishop score < 7 81 79 Postterm 65 69 Hypertension 18 12 Oligohydramnios 8 13 Other 11 10 Birth weight (grams) 3645 (566) 3585 (612)
Data given as number or mean (standard deviation)
Peripanum data are given in Table 4, with relative risk (R.R) and 95% Cl. There were
no significant differences in oxytocin (RR 0.67, CI 0.38, 1.18), epidural (R.R 0.89, CI 0.51, 1.56)
or other analgesia use (RR 0.59, CI 0.21, 1.64). No significant differences were observed in
meconium staining of amniotic fluid (RR 0.78, CI 0.41, 1.47), frequency of need for
episiotomies {RR 1.23, CI 0.62, 2.42) or tearing of the perineum (RR 0.98, CI 0.72, 2.33).
61
Table 4: Peripanum data
Vaginal Oral Relative 95°/o Confidence
(n = 102) (n = 104) Risk Interval Intrapartum _frequency_. Oxytocin use 59 70 0.67 0.38, 1.18 Epidural use 61 65 0.89 0.51, 1.56 No analgesia use 6 10 0.59 0.21, 1.64 Meconium 21 29 0.78 0.41, 1.47 Episiotomy 22 19 1.23 0.62, 2.42 Laceration 54 59 0.98 0.56, 1.69 Third or fourth degree
2 0 Laceration Intact perineunn 34 29 1.29 0.72, 2.33
Data for mean vaginal birth intervals are given in Table 5. There was a significant
difference in time to vaginal birth depending on route of administration. Those subjects
randomized to the vaginal misoprostol route had a significandy shorter time to delivery (846 ±
385 minutes versus 1072 ± 593 minutes in the oral misoprostol group, P = .004).
Furthermore, the induction-to-full dilatation interval was also significantly shorter in the
vaginal misoprostol group (763 ± 361 minutes versus 1003 ± 578 minutes in the oral
misoprostol group, P - .002). There were no statistically significant differences in the time of
first, second and third stages of labour between the two groups. This indicates that the
observed difference in time to delivery was not simply due to an increased time pushing during
the second stage of labour.
62
Table 5: Labour interval to vaginal birth
Vaginal Oral Misoprostol Misoprostol
Induction to vaginal birth 846 385) 1072 (593) Induction to full
763 061) 1003 (578) dilatation 1st stage of labour 342 204) 365 223) 2nu stage of labour 83 891 69 76) 3rd stage of labour 12 ( 14J 9.9 I 8.4)
Data given in minutes as mean (standard deviation)
~:· Student's t-test
Secondary Outcomes
Birth Route
p~·
.004
.002
.48
.26
.21
There was no significant difference in binh route between the two groups (x., ~ =
2.53, P - .47). With oral misoprostol, there were 58 spontaneous vaginal deliveries, 21
vacuum deliveries, 9 forceps-assisted deliveries and 16 cesarean births (15.3%}. With vaginal
misoprostol, there were 56 spontaneous vaginal deliveries, 15 vacuum deliveries, 8 forceps-
assisted deliveries and 23 cesarean binhs (22.5%). Non-reassuring FHR tracing was the
indication for two cesareans in the oral misoprostol group and six cesareans in the vaginal
misoprostol group (RR 2.47, CI 0.49, 12.49}.
Excessive Uterine Activity
Fetal heart rate and uterine activity tracing data were analyzed using non parametric
statistics based on number of masked physician reviewers making that classification of a given
63
tracing (Table 6). The inherent subjectivity involved in analyzing fetal heart graph and uterine
contraction monitoring strips led us to analyze the data based on agreement between
physicians. All four physician-reviewers blinded to group assignment agreed that 25 women in
the vaginal misoprostol group and 5 women in the oral misoprostol group e.xperienced
tachysystole (P - .01), and four women in the vaginal misoprostol group and no women in the
oral misoprostol group experienced hyperstimulation (P == .04).
Table 6: Classification of excessive uterine activity
0 1 2 3 4 Reviewers Reviewer Reviewers Reviewers Reviewers classified classified classified classified classified p ::· graph as: graph as: graph as: graph as: graph as:
Tracing classification Hypertonic > 90 sec
Oral misoprostol 1 8 15 26 54 0.68 Vaginal misoprostol 0 12 11 29 49
Hypertonic > 120 sec Oral misoprostol 7 22 24 20 31 0.57 Vaginal misoprostol 8 21 22 28 22
Tachysystole > 10 min Oral misoprostol 23 14 18 25 23 0.21 Vaginal misoprostol 16 15 18 21 31
Tachysystole > 20 min Oral misoprostol 52 21 16 10 5 0.01 Vaginal misoprostol 41 17 7 11 25
Hyperstimulation Oral misoprostol 84 14 5 1 0 0.04 Vaginal misoprostol 70 17 6 4 4
~- Mann-Whitney U test.
64
Although there was more frequent 20-minute tachysystole (P < .04) in the vaginal
misoprostol group, these results were not statistically significant by our preset level for
secondary analyses, but the trend warrants attention.
Neonatal Outcomes
Neonatal outcomes in both groups were similar (Table 7). Umbilical cord arterial
and venous blood acid-base analysis was performed on 89 of the 104 participants in the oral
misoprostol group and 92 of the 102 participants in the vaginal misoprostol group (90% of
neonates}. Two neonates from the vaginal misoprostol group had cord pH less than 7.00;
however, no neonate met the American College of Obstetricians and Gynecologists (ACOG)
criteria for birth asphyxia. ICll While it is reassuring that no neonate experienced an adverse
outcome, this study did not have sufficient power to draw conclusions about fetal safety. The
significance level for all secondary and hypothesis-generating analyses was set at P < .001 to
account for multiple testing.
65
Table 7: Neonatal outcome
Vaginal Oral Misoprostol Misoprostol
A COG criteria for birth N N asphyxia: Apgar 5 min < 3 0 0 CordpH < 7 2 0 Base deficit > 16 2 0
Cord blood: Mean Mean Cord pH 7.28 0.10 7.30 0.09 Base deficit 5.28 4.14 4.44 3.32
Apgar scores: 1\1edian Median Apgar 1 minute 9 (8,9) 9 8,9) Apgar 5 minute 9 (9,10) 9 ( 9,10)
N N Apgar 1 minute < 7 17 7 Apgar 5 minute < 7 0 1
::-Fishers' exact test
t Student's t test: mean (standard deviation)
+Mann-Whitney U test: median (1st and 3rd quartile)
x Chi square test
p
.16::-
.16::-
.08t
.13t
.59:!:
.23:!:
.03;{ .32::-
66
Labour Satisfaction
Ninety women in the oral group and 83 in the vaginal group completed a labour
satisfaction questionnaire administered 24 hours postpartum (84% response rate). Analysis of
the responses to all questions failed to detect any statistically significant difference in patient
satisfaction or overall score. All participants in the study reported positive labour and delivery
experiences. None of the panicipants reported adverse gastrointestinal effects. There was no
reported diarrhea in either group and no difference in occurrence of vomiting (18 in the oral
and 19 in the vaginal group, P - .81).
67
Cbapter 7
METHODOLOGICAL ISSUES AND DISCUSSION
Design Problems
The results of this study are convincing because of the strength of its masked
randomized design and the lack of protocol violations. The problems encountered in
conducting the trial stemmed from two major sources: financing the trial and accomplishing
blinding. From the point of view of financial support, it became apparent very early that the
Canadian manufacturer of misoprostol (Searle Canada, Oakville Ontario) was not interested in
supporting research on induction of labour. This related to medicolegal liability inherent to
research in pregnancy and a potential adverse effect on pharmaceutical marketing in the event
of publicized complications or lawsuits. Because of its reticence to be involved in pregnancy
related research, the manufacturer declined to provide a placebo. Estimates obtained from the
manufacturer placed the cost of placebo production at more than $200,000.
Even if misoprostol was identified to be an ideal pharmacological induction agent,
the market size and additional generated revenue would be small, as each potential patient
would require only small doses briefly, during an infrequent event in her lifetime. For these
reasons, the manufacturer is unlikely to ever be proactive in this area of research or to seek
Canadian Health Protection Branch approval of misoprostol for induction of labour. This
approval will have to be sought by investigators once more evidence is obtained
68
Since a major expected benefit of using misoprostol for labour induction is cost
savings, it was decided to pursue the possibility of realizing such a cost savings to obtain
financing for this trial from the Health Care Corporation of St. John's. A review of pharmacy
records from previous misoprostol studies conducted at the Grace General Hospital provided
data to suppon an expected savings by using misoprostol instead of dinoprostone for labour
induction, justifying financial suppon of this trial by the Health Care Corporation.
The Corporation approved the hiring of a part-time pharmacy technician to work
exclusively on the trial to provide a placebo for the study and to maintain records, ensuring
that each panicipant received the appropriate treatment. This technician was responsible for
the preparation of code-labelled indistinguishable drug-placebo packages according to the
specifications of the study protocol. Study drug packages contained eight doses of the
designated treatment, as the ma.ximum number of doses required by any subject in our prior
research studies was seven. The induction method was concealed from all caregivers and
investigators until data analysis was completed.
The Health Care Corporation also approved the hiring of a pan-time research nurse
to provide suppon to the Department of Obstetrics and Gynecology for the duration of the
trial. The research nurse reviewed all aspects of the study protocol and acted as a liaison with
the pharmacy department. Educational sessions were arranged for prenatal instructors,
residents and attending physicians to familiarize all participating caregivers with the study
protocol. Data collection, including the administration of a postpartum labour satisfaction
questionnaire to all participants, was the primary responsibility of the research nurse.
69
In addition to financial suppon from the Health Care Corporation, funding was
sought from the Medical Research Council of Canada. Two separate grant applications were
submitted; unfortunately, neither application was successful. An application to the Memorial
University of Newfoundland Faculty of Medicine was successful in securing a $25,000 research
grant, and was awarded the Cox Research Award. These additional funds were instrumental in
financing the remainder of the projected costs of the trial. Efforts to secure financing for this
trial delayed its stan date by approximately one year.
Recruitment
During the study period, 393 women presented for induction of labour. Of this
group, 85 were ineligible according to the study protocol, 17 did not wish to panicipate in a
study and 85 were not invited because the attending physician chose alternate methods of
induction. Two hundred and six of the 223 women (> 90%) invited to participate agreed to
panicipate in the study. Since only 206 of a potential308 were recruited, a possible selection
bias is introduced into the study. Non-participants were documented but their demographic
characteristics were not. This is because physicians other than the investigators were
providing primary medical care. We did not have permission from non-panicipants to review
their medical records.
Differences are widely acknowledged to exist between patients who agree to
participate as subjects in research studies and those who do not. Significantly different
outcomes have frequently been documented even between non-participants in studies and
participants randomized to an equivalent control arm. Therefore, one must interpret these
70
results (and those of most trials) bearing in mind that generalizability to the clinical population
may not be complete. In this case, 85 patients were not invited to participate in the study
because their attending physician preferred alternate methods of induction including
amniotomy, oxytocin or dinoprostone. It is likely that several of these women had a
favourable cervix (high Bishop score) at presentation, leading the attending physician to
believe that an amniotomy followed by oxytocin was a more appropriate treatment than
randomization to either form of misoprostol induction. Thus, any conclusions reached
concerning the relative merits of the forms of misoprostol induction studied here must be
understood to apply to a relatively similar population to that randomized.
Classification of Excessive Uterine Activity
Although there is no direct relationship between continuous electronic fetal
monitoring and fetal well-being, non-reassuring fetal heart rate changes associated with
excessive uterine activity warrant intervention either to reduce uterine activity or to deliver the
fetus. Because of confusion stemming from inconsistent definitions of excessive uterine
activity, Curtis recommended a standard terminology. While it allows for classification of
labour graphs, the terminology is nevertheless arbitrary, particularly with respect to
tachysystole and hypertonic contractions.
Despite concerns of possible uterine hyperstimulation, substantive adverse newborn
outcomes such as neonatal acidosis, meconium aspiration, and birth asphyxia have been rare
and have not differed between misoprostol and control groups in previous published studies.
In our study, two neonates in the vaginal misoprostol group had cord blood pH less than 7.00
71
but no asphyxia occurred. Among the 39 cesarean births, six of the eight done for fetal safety
reasons were in the vaginal misoprostol group. While these differences ace not statistically
significant, the question is clinically relevant, making it a worthwhile primary goal in the design
of a future clinical trial. The frequency of excessive uterine activity associated with route of
administration was evaluated as a secondary outcome in the present study, and did not form a
basis for calculation of sample size.
Inherent in the evaluation of excessive uterine activity is the concern about inter
observer and intra-observer variation. We believe that agreement about the occurrence of
hyperstimulation among all four reviewers blinded to each others' results and to group
assignment provides strong evidence for the e.xistence of four cases of hyperstimulation in the
vaginal misoprostol group and no cases of hyperstimulation in the oral group. The finding
that two neonates in the vaginal misoprostol group that had a cord pH less than 7.00 further
supports the conclusion that vaginal misoprostol has a greater propensity to cause excessive
stimulation of uterine contractility.
Pharmacokinetics of Crushed Misoprostol
The pharmacokinetics of misoprostol suspended in gel and given vaginally has not
been directly evaluated. No study has evaluated the pharmacokinetics of crushed misoprostol
given orally. In our study, this group had the lowest incidence of tachysystole and
hyperstimulation. Understanding the timing of e."<cessive uterine activity is important because
it addresses concerns about fetal safety. Appropriate and timely intervention can then be
72
initiated. Such information is also essential for the rational design of further trials to determine
the safest dose, form, and route of misoprostol administration.
A Future Study of Excessive Uterine Activity
To further study the relationship between misoprostol and excessive uterine activity,
our research group has planned a cohort study. The primary objective is to determine the
incidence and timing of excessive uterine activity with induction of labour with misoprostol.
Different routes and formulations of misoprostol will be compared to prostaglandin E1
intracervical or intravaginal gel, oxytocin, and spontaneous labour.
Study Design
The study will include women enrolled in each of three randomized trials evaluating
misoprostol for labour induction. tct,tOJ, tcs The studies were conducted from March 30, 1994 to
September 22, 1994 and October 25, 1996 to December 19, 1997 at the Grace General
Hospital in St. John's, Newfoundland. A cohort of women who presented in spontaneous
labour during the same period will also be evaluated.
Inclusion criteria are: a single live fetus in cephalic presentation, gestation longer than
37 weeks and fetal heart rate and uterine activity graphs available for review. Exclusion criteria
include: non-reassuring fetal heart rate tracing prior to induction (or on admission in the
spontaneous labour cohort), prior uterine surgery, contraindication to vaginal birth, age less
than 19 years and known hypersensitivity to misoprostol or other prostaglandins.
73
One of these trials enrolled only women with term premature rupture of
membranes106, while the other two trials included only those women with intact
membranes. 101'1ru The details of the three trials are previously published. For those women
undergoing induction oflabour, the cohorts will be as follows: misoprostolSO ~tg crushed and
suspended in hydroxyethyl gel given vaginally every four hours, misoprostolSO ~-&g crushed and
mixed with juice given orally every four hours, misoprostolSO ~-&g tablet given orally every four
hours for premature rupture of membranes [PROM], intravenous oxytocin beginning at 2
milliunits (mu) I min and increasing by 2 mu/ min every 15 to 30 minutes for PROM,
misoprostol tablet 50 ~-&g vaginally every four hours, prostaglandin E! intracervical gel 0.5 mg
every six hours or prostaglandin E2 intravaginal gell mg or 2 mg every 6 hours. The cohort of
women in spontaneous labour in the study will include a group of women who had at least si.x
hours of fetal heart rate and uterine activity monitoring and who would have otherwise met
the criteria for induction of labour.
Methods
The fetal heart rate and uterine activity graphs will be independendy and individually
analyzed by three maternal fetal medicine specialists and one physician enrolled in a Maternal
Fetal Medicine Fellowship. Tachysystole will be defmed as a contraction frequency of more
than five in a ten-minute period for two consecutive ten-minute periods. Hyperstimulation
will be defined as tachysystole with late fetal he-.ut rate decelerations or fetal tachycardia greater
than 160 beats per minute. Tachysystole or hyperstimulation will be noted if at least two of
those reviewing the graph agree with the diagnosis.
74
For those women receiving the crushed forms of misoprostol, reviewers were
blinded to group usignment and the code was not broken until completion of analysis of
graphs. For the other cohorts, although the reviewers were not blinded, they will not actively
seek to reveal the cohon allocation until analysis of graphs is completed. The timing of the
excessive uterine activity will be determined relative to the misoprostol dosing schedule. If
tachysystole or hyperstimulation occurred after subsequent doses of misoprostol, the timing of
each episode will be determined relative to the number of doses given. Those women who
experience tachysystole or hyperstimulation only after oxytocin augmentation is initiated, or
during the second stage of labour with pushing, will not be included in the determination of
timing of excessive uterine activity, as this may lead to confounding.
Sample Size Calculation
The sample size required is based on the estimated incidence of tachysystole in the
misoprostol and oxytocin cohorts. 101'103
'1c6 Using a two-tailed a = .05 and ~ = .20. fifty-one
women in each cohon will be needed to detect a significant difference between 35% incidence
in one cohon and a 10% incidence in another cohon. Data will be analyzed using Statisti.x 4.1
(Analytical Software, Tallahassee, FL). Continuous variables will be assessed using Student's t
test and analysis of variance, while categorical variables will be analyzed by "land Fisher e.xact
test where appropriate. Statistical significance will be set at P < .05. Because of multiple
testing, all comparisons other than the primary outcomes will be considered only as hypothesis
generaung.
75
Chapter 8
IS ORAL MISOPROSTOL THE IDEAL INDUCTION AGENT?
Obstetricians routinely use various mechanical methods and pharmacological agents
to induce labour, however, no single approach has been universally successful. Oxytocin is an
effective induction agent; nonetheless, it is not optimal therapy for a patient with an unripe
cervix. Administration requires an intravenous infusion with continuous monitoring of fetal
heart rate and uterine contractions. Prostaglandins are effective cervical ripening agents,
achieving a shorter induction-to-delivery interval and a lower rate of operative intervention
when compared to oxytocin.
Oral administration of prostaglandins has essentially been abandoned because of
undesirable gastrointestinal effects, which have limited their use to vaginal and intracervical
routes. Termination of prostaglandin-related e."'{cessive uterine activity remains problematic.
An oral induction agent that is safe, effective, inexpensive and well tolerated by patients would
likely be attractive to patients and health care providers. Anticipated benefits include
avoidance of intravenous lines in some panurients, less frequent need for vaginal e."'{amination
and greater freedom for upright positioning. Ambulation might even facilitate labour progress.
Misoprostol
Misoprostol (Cytotec®; Searle, Oakville, ON), a synthetic prostaglandin (PGEJ
analogue, is marketed in an oral formulation of 100 f.lg or 200 f.lg tablets.82 It is currently
76
approved by the U.S. Food and Drug Administration for use in the treatment and prevention
of gastrointestinal ulcer disease resulting from non-steroidal anti-inflammatory use.
Misoprostol is an inexpensive prostaglandin capable of initiating uterine contractions;
therefore, it has been exten.sively investigated for use as an induction agent in pregnancy. A
single 100 J,lg tablet costs$ 0.30; therefore, a single 50 J,lg dose may cost as little as $0.15.
Misoprostol is also temperature-stable and does not have special storage requirements.
Pharmacokinetics of Oral and Vaginal Misoprostol
Misoprostol is primarily metabolized in the liver, with less than one percent of its
active metabolite excreted in urine. Misoprostol has no known drug interactions and does not
induce the hepatic cytochrome P-450 enzyme system. The most common adverse effects of
misoprostol are nausea, vomiting, diarrhea, abdominal pain, chills, shivering and fever, all of
which are dose dependent. Although other prostaglandins (prostaglandin E 2 and
prostaglandin F 2a) can cause myocardial infarction and bronchospasm, misoprostol does not.
Toxic doses of misoprostol have not been determined.
Zieman et al. 110 compared the pharmacokinetics of misoprostol (400 J.tg)
administered by vaginal and oral routes to 10 non-pregnant and 10 pregnant first trimester
volunteers, concluding that significant differences exist. Among women who were 9 to 11
weeks pregnant and given misoprostol before a surgical abortion, intrauterine pressure began
to increase an average of 8 minutes after oral administration and 21 minutes after vaginal
administration and was maximal 25 minutes after oral administration and 46 minutes after
vaginal administration. Uterine contractility initially increased and reached a plateau one hour
77
after oral administration whereas uterine contractility increased continuously for four hours
after vaginal administration. Maximal uterine contractility was significandy higher after vaginal
administration. Systemic bioavailability of vaginally administered misoprostol was found to be
three times that of misoprostol administered orally. With vaginal administration, peak plasma
levels were reached more slowly but were sustained in excess of four hours.
In a confirmatory study, Danielsson et al. 111 reponed that uterine activity increased
continuously for four hours after vaginal administration of misoprostol, and warned that this
could lead to a cumulative effect with subsequent dosing. When contractions of abnormally
high intensity or frequency occur, insufficient recovery resulting in FHR abnormalities could
necessitate operative delivery or lead to fetal asphyxia. It could reasonably be argued that the
greater frequency of abnormal fetal bean rate patterns observed in patients who receive vaginal
misoprostol are a result of excessive uterine activity, reflective of this greater bioavailability.
Induction agents, ideally, should mimic spontaneous labour while avoiding e."~Ccessive
uterine activity. Inherent in the evaluation of excessive uterine activity is the concern about
inter-observer and intra-observer variation. We believe that agreement about the occurrence
of hyperstimulation among all four reviewers blinded to each others' results and to group
assignment provides strong evidence for the existence in this study of four cases of
hyperstimulation in the vaginal misoprostol group and no cases ofhyperstimulation in the oral
group. The finding that two neonates in the vaginal misoprostol group that had a cord pH less
than 7.00 further supports the conclusion that vaginal misoprostol has a greater propensity to
cause e."~Ccessive stimulation of uterine contractility. Uterine hyperstimulation was a secondary
outcome and was not considered in calculation of sample size. The significance level for all
78
secondary and hypothesis-generating analyses was set at P < .00 l to account for multiple
testing.
Misoprostoi-Current and Future Use
Despite accumulating evidence of effectiveness and obvious cost advantage in over
44 randomized clinical trials, the U.S. Food and Drug Administration and the Therapeutic
Products Program of Health Canada have not yet approved misoprostol for induction of
labour. However, they do recognize that in certain circumstances, off-label uses of approved
products are appropriate, rational and accepted medical practice. Misoprostol administered
vaginally for labour induction is currently in widespread use in the United States. Its use in
' most Canadian tertiary care centres is limited to research protocols.
The American College of Obstetricians and Gynecologists (ACOG) Committee on
Obstetric Practice recently concluded that vaginal misoprostol administered appears to be
effective in inducing labour in pregnant women who have unfavourable cervi.'Ces; however, the
initial dose should be limited to 25 !Jg every three to fours hours because of concerns of the
greater incidence of tachysystole observed with the use of higher doses. It further concluded
that prospective studies are necessary to define optimal dosing regimens. 111 The Society of
Obstetricians and Gynecologists of Canada (SOGC) has not published a recommendation on
the use of misoprostol as an induction agent.
Misoprostol is an effective induction agent whether it is given orally or vaginally.
Misoprostol is much less expensive than dinoprostone and, unlike dinoprostone, which
79
reqwres refrigeration until just before use, it is stable at room temperature. Oral
administration of misoprostol is appealing for several reasons, including convenience and a
non-invasive mode of delivery. Fewer cervical exams could also potentially result in lower
peripartum infection rates, particularly in women with prelabour-ruptured membranes. Oral
administration of misoprostol, if proved s~e and effective, could potentially reduce overall
hospitalization time by permitting administration of the medication in an outpatient setting.
No studies using misoprostol as an outpatient preinduction cervical ripening agent have been
published. Use for this indication is not currently recommended because of the need for dose
surveillance of uterine activity after administration of misoprostol. In addition to these
potential benefits, oral misoprostol shares with vaginal misoprostol the advantages of cost
effectiveness and efficacy in labour induction.
The oral route of misoprostol administration is an important one that has not been
fully investigated, and the question of whether oral may be the preferred route of misoprostol
administration remains to be answered. Sanchez-Ramos et al. 99 conducted a systematic review
of published randomized trials comparing oral and vaginal misoprostol administration for
cervical ripening and labour induction. In total, 1,191 patients were randomized to receive
misoprostol orally (n - 602) or by the vaginal route (n - 589). The oral doses ranged from 50
J.lg to 200 J.tg every four to six hours. Vaginal misoprostol was administered in doses ranging
from 25 f,lg every four hours to 100 J.tg every three hours.
No significant difference was noted in the proportion of patients who delivered
within 12 and 24 hours. Similarly, the intervals from start of induction to vaginal delivery were
80
not different. No difference was noted in the incidence of abnormal Apgar scores or rates of
admission to neonatal intensive care units. Interestingly, the rate of cesarean delivery was
significantly lower in the oral misoprostol group (15% versus 22%, Mantel-Haenszel pooled
odds ratio 0.64).
Although meta-analysis provides a systematic and explicit method for demonstrating
early evidence with regard to the effectiveness of treatments, it is not as valuable as a large
randomized trial. Three principal concerns must be addressed when assessing the validity of a
meta-analysis: publication bias, quality assessment of the randomized clinical trials included
and study heterogeneity. Several sources of inter-study heterogeneity can be identified in the
Sanchez-Ramos review. There were variations in treatment protocols, small sample size in two
of the seven studiestco,tct and a lack of uniform stratification according to Bishop score and
parity. It is imponant that the results of a meta-analysis not depend excessively on the results
of a single study. To assess inter-study heterogeneity, the analysis should be repeated after
sequentially excluding each trial, seeking to determine whether any one trial contributed
excessively to the overall reduction in cesarean delivery rate.
The enormous economic and clinical advantages of misoprostol for cervical ripening
and labour induction suggest that the unanswered questions about the impact of oral
misoprostol on cesarean delivery rates and fetal safety should be addressed in a properly
designed, well-conducted multi-centre trial. Our research group has planned such a trial, and is
currently seeking funding to allow it to proceed.
81
REFERENCES
1. Bishop EH. Pelvis scoring for elective induction. Obstet Gynecol1964; 24:266-8.
2. Beasdale JE, Johnston JM: Prostaglandins and human parturition: Regulation of
arachidonic acid mobilization. Rev Perinat Med 5:151,1984.
3. Honican D, Brockman DE, Myatt L Transcriptional and post-translational regulation
of cytosolic phospholipase A 2 by 11.-1 and EGF in WISH cells. J Soc Gynecol Invest.
5:59A, 1998.
4. Rhee SG, Choi KD. Multiple isoforms of phospholipase C isoenzymes and their
activation mechanisms. Adv.llnd Mess. Phospho. Res.26:35-61,1992.
5. Simmons DL, Levy Y, T annoni, Erikson. Identification of a phorbol-ester-repressible
v-src-inducible gene. Proc. Nad. Acad. Sci. USA 86:35-61,1992.
6. Dewitt DL, Meade E, Smith WL. PGH synthase isoenzyme selectivity: The potential
for safer nonsteroidal anti-inflammatory drugs. AmJ Med. 95:40S-44S, 1993.
7. Smith WL, Dewitt DL. Prostaglandin endoperoxide H synthases-1 and 2. Adv.
Immunol. 62: 167-215,1996.
8. Herschman HR. Prostaglandin synthase 2. Biochim. Biophys. Acta 1299: 125-40,1996.
9. Mijovic JE, Zakar T, N aim TK, Olson DM. Prostaglandin endoperoxide H synthase
( PGHS) activity increases with term labour in human chorion. Am J Physiol. 272:
E832-E840, 1997.
10. Mijovic JE, Zakar T, Naim TK, Olson DM .. Prostaglandin endoperoxide H synthase
( PGHS) activity and PGHS-1 and PGHS-2 messenger ribonucleic acid abundance in
82
human chorion throughout gestation and with preterm labour. J Clio Endocrinol
Metab 83: 1358-67, 1998.
11. Hirst D, Mijovic JE, Zakar T, Olson OM. Prostaglandin endoperoxide H synthase-!
and 2 m RNA levels and enzyme activity in human deciduas at term labour. J Soc
Gynecol Invest, 5:13-20,1998.
12. Olson OM, Zakar T, Mitchell. Prostaglandin synthesis regulation by intrauterine
tissues. In Molecular Aspects of Placental and Fetal Membrane Autocoids. Rice GE,
Brennecke SP. (eds) Boca Raton, Fl,CRC Press 1993, 55-95.
13. Negishi M, Sugimoto Y, Ichikawa A. Molecular mechanisms of adverse actions of
prostanoid receptors. Biochim Biophys Acta 1259:109,1983
14. Narumiya S, Sugimoto Y, Ushikubi F. Prostanoids receptors: structure, properties and
functions. Physiol Rev. 1999;79:1193-1226.
15. Brodt-Eppley J, Myatt L: Expression of contractile FP and relaxatory EP ~ receptors in
human myometrium during gestation and with labour. J Soc Gynecol Invest
S(Suppl): 133,1998.
16. Breyer RM, Davis LS, Nian C et al. Cloning and expression of the rabbit prostaglandin
EP,. receptor. AmJ Physiol.1996;270:F485-F493
17. Challis JRG, Gibb W. Control of parturition. Prenat Neonatal Med 1:283,1996
18. van Meir CA, Sangha RK., Walton JC, Mathews SG, Keirse MJNC, Challis JRG:
Immunoreactive 15-hydroxyprostaglandin dehydrogenase {PGOH) is reduced in fetal
membranes from patients at preterm delivery in the presence of infection. Placenta
17:291, 1996.
83
19. van Meir CA, Mathews SG, Keirse MJNC, Ramirez !viM, Bocking A, Challis JRG: 15-
hydroxyprostaglandin dehydrogenase (PGDH}: Implications in preterm labour with
and without ascending infection. J Clin Endocrinol Metab 82:969,1997.
20. van Meir CA, Mathews SG, Ramirez MM, Calder AA, Keirse MJNC, Challis JRG:
Chorionic prostaglandin catabolism is decreased in the lower uterine segment with
term labour. Placenta 18:109, 1997b.
21. Mathews SG, ChallisJRG: Regulation of the hypothalamic-pituatary-adrenal axis in
late gestation. Trends Endocrinol metab 7:14,1996
22. Lye SJ. Initiation of parturition. Anim Reprod Sci 42:495,1997.
23. Robinson BG, Emanuel RL, Frim DM,Majzoub JA.: Glucocorticoid stimulates
expression of corticotrophin-releasing hormone gene in human placenta. Proc N ad
Acad Sci USA 85:5244,1988.
24. Jones SA, Brooks AN, Challis JRG: Steroids modulate corticotrophin- releasing factor
production in human fetal membranes and placenta. J Clin Endocrinol Metab
68:825,1989.
25. Jones SA, Challis JRG. Local stimulation of prostaglandin production by
corticotropin-releasing hormone in human fetal membranes and placenta. Biochem
Biophys Res Commun, 1989; 159: 192-9
26. Jones SA, ChallisJRG. Steroid, corticotrophin-releasing hormone ACTH and
prostaglandin interaction in the amnion and placenta of early pregnancy in man. J
Endocrinol 1990; 125:153-9
27~ Zakar T, Olson D: Dexamethasone stimulates arachidonic acid conversion to
prostaglandin E2 in human amnion cells. J Dev Physiol 12:269,1989.
84
28. Quantero HW. Placental Corticotrophin releasing factor may modulate human
parturition. Placenta 1989;10:439-43.
29. Petraglia F, Florio P, Nappi C, Genazzani AR: Peptide signalling in human placenta
and membranes. Autocrine,paracrine and endocrine mechanisms. Endocr Rev
17: 156,1996.
30. Danforth ON: The fibrous nature of the human cervi.'IC, and its relation to the isthmic
segment in gravid and nongravid uteri. AmJ Obstet Gynecol53:541,1947.
31. Liggins GC. Cervical ripening as an inflammatory reaction. In: Ellwood DA and
Anderson ABM (eds). The cervix in Pregnacy and Labour: Clinical and Biochemical
lnvestgiations. Edinburgh, UK: Churchill Livingstone,1981.
32. Uldbjerg N, Ekman G, Malstrom A, Olsson K, Ulmsten U. Ripening of the human
uterine cervix related to changes in collagenolytic activity. Am J Obstet Gynecol.
1983;147:662-6.
33. Salvemini D, Misko TP, Masferrer JL, Seibert K, Currie MG, Needleman P. Nitric
oxide activates cyclooxygenase enzymes. Proc Nat Acad Sci USA. 1993;90:7240-44
34. Ellwood DA, Mitchell MD, Anderson ABM, Turnbull AC. The invitro production of
prostanoids by the human cervix during pregnancy: Preliminary observations. Br. J
Obstet Gyneaecol. 1980;87:210-14.
35. Norstom A: Acute effects of prostaglandin on the biosynthesis of connective tissue
constituents in the nonpregnant human cervix uteri. Acta Obstet Gynecol Scand,
1984;63:169.
85
36. Murota S, Abe M, Otsuka K. Stimulatory effect of prostaglandins on the production of
hexosamine-containg substances by cultured fibroblasts. Prostaglandins. 1977; 14:983-
91.
37. Thiery M, Baines CJ, Keirse MJ. The development of methods for inducing labour.
In: Chalmers I, Enkin MW, Keirse MJ, editors. Effective Care in Pregnancy and
Childbirth. Oxford: Oxford University Press, 1989:970.
38. McColgin SW, Hampton HL, McCaulJF, Howard PR, Andrew ME, MorrisonJC.
Stripping of membranes at term: can it safely reduce the incidence of post-term
pregnancies? Obstet Gynecoll990; 76:678-80.
39. Boulvain M, Irion 0. Stripping/sweeping the membranes for inducing labour or
preventing post-term pregnancy (Cochrane Review). In: The Cochrane Library, Issue
4, 2000. Oxford: Update Software.
40. Swann R. Induction of labor by stripping membranes. Obstet Gynecoll958; 11:7 4-8.
41. McColgin SW, Bennett W A, Roach H. Parturitional factors associated with membrane
stripping. Am J Obstet Gynecol 1993; 169:71.
42. Weissberg SM, Spellacy WN. Membrane stripping to induce labour. J Reprod Med
1977; 19{3):125-7.
43. Allot HA, Palmer CR. Sweeping the membranes: A valid procedure in stimulating the
onset of labour. Br J Obstet Gynaecoll993; 100:898.
44. El-Torkey M, Grant JM. Sweeping of the membranes is an effective method of
induction of labour in prolonged pregnancy. A randomized trial. Br. J Obstet
Gynecol1992; 99:455-8.
86
45. Crane J, Bennett KA, Windrim R, Kravitz H, Young DC. The effectiveness of
sweeping membranes at term: A randomized trial. Obstet Gynecol1997; 89:586-90.
46. Cammu H, Haitsma V. Sweeping of the membranes at 39 weeks in nulliparous
women: A randomized controlled trial. Br 1 Obstet Gynaecol1998; 105:41-4
47. Boulvain M, Fraser WD, Marcoux S, Fontaine JY, Bazin S, Pinault JJ et al. Does
sweeping of the membranes reduce the need for formal induction of labour? A
randomized controlled trial. Br 1 Obstet Gynaecol1998; 105(1):34-40.
48. Wiriyasirivaj B, V utayavanich T, Ruangsri R. A randomized trial of membrane
stripping at term to promote labour. Obstet Gynecoll996; 87:767-770.
49. Berghella V, Rodgers RA, Lescale K. Stripping of membranes as a safe method to
reduce prolonged pregnancies. Obstet Gynecol 1996; 87:927
50. Barkai G, Cohen SB, Kees. Induction of labor with use of a Foley catheter and extra
amniotic corticosteroids. AmJ Obstet Gynecol1997; 177:1145.
51. Thomas IL, ChenowethJN, Trone GN. Preparation for induction of labor of the
unfavorable cervix with Foley catheter compared with vaginal prostaglandin. Aust NZ
J Obstet Gynecol 1986; 26:30.
52. Sherman DJ, Frenkel E, TovbinJ et al. Ripening of the unfavorable cervix with e.xtra
amniotic catheter balloon: Clinical experience and review. Obstet Gynecol Surv 1996;
10:621.
53. Hemlin J, Moller B. Extra-amniotic saline infusion is promising in preparing the cervi.x
for induction of labor. Acta Obstet Gynecol Scand 1998; 77:45.
87
54. Schreyer P, Sherman D J, Ariely S. Ripening the highly unfavorable cervix with extra
amniotic saline instillation or vaginal prostaglandin E2 application. Obstet Gynecol
1989; 73:938.
55. Rouben D, Arias F. A randomized trial of extra-amniotic saline infusion plus
intracervical Foley catheter balloon versus prostaglandin E1 vaginal gel for ripening the
cervix and inducing labor in patients with unfavorable cervices. Obstet Gynecol1993;
182:290.
56. L yndrup J, Nicleson C, Weber T. Induction of labor by balloon catheter with extra
amniotic saline infusion (BCEAS): A randomized comparison with PGE1 vaginal
pessaries. Eur J Obstet Gynecol Reprod Biol1994; 53:189.
57. Keirse MJ, Thiery M, Parewijck W. Chronic stimulation of uterine prostaglandin
synthesis during cervical ripening before the onset of labor. Prostaglandins 1983;
25:671.
58. Ezimokhai M, NwabineliJN. The use of Foley's catheter in ripening the unfavorable
cervix prior to induction of labor. Br J Obstet Gynaecol1980; 87:281.
59. Lewis GL. Cervical ripening before induction of labor with prostaglandin E1 pessaries
or a Foley's catheter. J Obstet Gynecol1983; 3:173.
60. AtadJ, Hallak M, Auslender R. A randomized comparison of prostaglandin E1,
oxytocin, and the double-balloon device in inducing labor. Obstet Gynecol 1996;
87:223.
61. St. Onge R, Connors G. Preinduction cervical ripening: A comparison of intracervical
prostaglandin E1 versus the Foley catheter. AmJ Obstet Gynecol1995; 172:687-90.
88
62. Chamberlain JE, Natale R, Schmuck ML. Efficacy and patient satisfaction with
outpatient prostaglandin E2 versus Foley catheter for cervical ripening- A randomised
trial. Journal SOGC 1996; 1094-9.
63. Cross WG, Pitkin RM. uminaria as an adjunct in induction oflabor. Obstet Gynecol
1978; 51:606.
64. Gower RH, Toraya J, Miller JM. Laminaria for preinduction cervic-al ripening. Obstet
Gynecol1982; 60:617.
65. Rosenberg LS, Tejani NA, V aranasi M. Preinduction ripening of the cervi.x with
laminaria in the nulliparous patient. J Reprod Med 1980; 25:60.
66. T ohan N, Tejani NA, V aranasi M. Ripening of the term cervi.x with laminaria. Obstet
Gynecol 1979; 54:588.
67. Lin A, Kupferminc M, Dooley SL. A Randomized trial of extra-amniotic saline
infusion versus laminaria for cervical ripening. Obstet Gynecol1995; 86:545.
68. Krammer J, Williams MC, Sawai SK. Pre-induction cervical ripening: A Randomized
comparison of two methods. Obstet Gynecol1995; 85:614.
69. Sanchez-Ramos L, Kaunitz AM, Connor PM. Hygroscopic cervical dilators and
prostaglandin E2 gel for preinduction cervical ripening: A randomized prospective
comparison. J Reprod Med 1992; 37:355.
70. Medvei VC. A history of clinical endocrinology. Medical and Technical Press.
Lancaster: 1982,722-3.
71. Bell WB. The pituitary body and the value of infundibular extract in shock, uterine
atony and intestinal paresis. Br MedJ 1909; 1609-13.
89
72. Embrey MP. Induction of labour with prostaglandins E1 and E2• Br MedJ 1970;
2:256-8.
73. Rayburn WF. Clinical experience with a controlled-release prostaglandin E2 in the
USA. Br J Obstet Gynaecol1997; 104 supplement 15.
7 4. Rayburn WF, Wapner RJ, Barss VA, Spitz berg E, Molina RD, Mandsager Net al. An
intravaginal controlled-release prostaglandin E2 pessary for cervical ripening and
initiation of labour at term. Obstet Gynecol 1992; 79(3):374-9.
75. Keirse MJ. Any prostaglandin, any route for cervical ripening. In the Cochrane
Pregnancy and Childbirth Database 1995, Issue 2.
76. Keirse MJ, de Koning Gans HJ, for the Dutch Collaborative Prostaglandin Trialists
Group. Randomized comparison of the effects of endocervical and vaginal
prostaglandin E2 gel in women with various degrees of cervical ripeness. Am J Obstet
Gynecol1995; 173:1859-64.
77. Witter FR, Rocco LE, Johnson TRB. A Randomized trial of prostaglandin E2 in a
controlled-release vaginal pessary for cervical ripening at term. AmJ Obstet Gynecol
1992; 166:830-4.
78. Witter FR, Mercer BM. Improved intravaginal controlled-release prostaglandin E2
insert for cervical ripening at term. The Prostaglandin E2 Insert Study Group. J
Matern Fetal Med 1996; 5{2):64-9.
79. Chyu JK, Strassner HT. Prostaglandin E2 for cervical ripening: A Randomized
comparison of Cervidil versus Prepidil. Am J Obstet Gynecol 1997; 177:606-11.
90
80. Crane JM, Bennett K.A. A meta-analysis of controlled- release prostaglandin for
cervical ripening and labour induction. J Soc Obstet Gynecol2000 Sept; 22:692-8.
81. American College of Obstetricians and Gynecologists Committee opinion.
Monitoring during induction of labour with dinoprostone. 1998; 209:135.
82. Garris RE, Kirkwood CF. Misoprostol: A prostaglandin E1 analogue. Clinical
Pharmacy 1989; 8:627-44.
83. Margulies M, Perez GC, Voto LS. Misoprostol to induce labour. (letter). Lancet 1992;
339:64.
84. Sanchez-Ramos L, Kaunitz AM, Del Vaile GO, Oelke I, Schroeder PA, Briones DK.
Labor induction with the prostaglandin E, methyl analogue misoprostol versus
oxytocin: A randomized trial. Obstet Gynecol 1993; 81:332-6.
85. Wing DA, Jones MM, Rahall A, Goodwin TM, Paul RH. A comparison of
misoprostol and prostaglandin E2 gel for preinduction cervical ripening and labor
induction. AmJ Obstet Gynecoll995; 172:1804-10.
86. Mundie WR, Young DC. Vaginal misoprostol for induction of labor: A randomized
controlled trial. Obstet Gynecol1996; 88,521-5.
87. Wing DA, Rahall A, Jones MM, Goodwin TM, Paul RH. Misoprostol an effective
agent for cervical ripening and labour induction. Am J Obstet Gynecol 1995;
172:1811-6
88. Sanchez-Ramos, Kaunitz A, Wears R, Oelke I, Gaudier F. Misoprostol for cervical
ripening and labor induction: A meta-analysis. Obstet Gynecol1997; 89:633-41.
91
89. Fletcher HM, MitchellS, Simeon D, Frederick], Brown D. Intravaginal misoprostol as
a cervical ripening agent. Br J Obstet Gynecol1993;100:641-4
90. Fletcher HM, MitchellS, Simeon D, Frederick], Brown D.lntravaginal misoprostol
versus dinoprostone as cervical ripening and labour inducing agents. Obstet Gynecol
1994;83:244-7.
91. Campos GA, Guzman S, RodriguezJG, Voto LS, Marguiles M. Misoprostol-Un
analog de Ia PGE 1 - para Ia induccion de parto a termino : Estudio comparative y
randomizado con oxitocina. RevChil Obstet Ginecol 1994;59: 190-6.
92. Varaklis K, Gumina R, Stubblefield PG. Randomized controlled trial of vaginal
misoprostol and prostaglandin E :! gel for induction of labour at term. Obstet Gynecol
1995;86:541-4.
93. Chuck FJ, Huffaker BJ. Labour induction with intracervical prostaglandin E! gel: A
Randomized comparison. AmJ Obstet Gynecol1995;173:1137-42.
94. Wing DA, Paul RH. A comparison of differing dosing regimes of vaginally
administered misoprostol for preinduction cervical ripening and labor induction. AmJ
Obstet Gynecoll996; 175:158-64.
95. Sanchez-Ramos L, Peterson DE, Oelke I, Gaudier FL, Kaunitz AM. Labour
induction with prostaglandin E1 misoprostol compared with dinoprostone vaginal
insert: a randomized trial. Obstet Gynecol1998; 91:401-5.
96. Wing DA, Ortiz-Omphroy G, Paul RH. A comparison of intermittent vaginal
administration of misoprostol with continuous dinoprostone for cervical ripening and
labour induction. Am J Obstet Gynecol1997; 177:612-8.
92
97. Hoffmeyr GJ, Gulmezoglu Am, Alfievic. Misoprostol for induction of labour: A
syste1natic review. Br J Obstet Gynaecol1999;798-803.
98. Keirse MINC. Prostaglandins in preinduction cervical ripening :meta-analysis of
woldwide clinical experience. J Reprod Med 1993;38(suppl}:89-
99. Sanchez-Ramos, Kaunitz M. Misoprostol for cervical ripening and labour induction: A
systematic review of the literature. Clinical Obstetrics and Gynecology, Vol43. No.3
100. Ngai SW, To WK, Lao T, Ho PK. Cervical priming with oral misoprostol in pre-labor
rupture of membranes at term. Obstet Gynecol 1996; 87:923-6.
101. Windrim R, Bennett KA, Mundie W, Young DC. Oral administration of misoprostol
for labor induction: A randomized controlled trial. Obstet Gynecol1997; 89(3):392-
7.
102. Toppozada MK, Anwar MY, Hassan HA, El-Gazaerly WS. Oral or vaginal
misoprostol for induction of labor. Int J Gynaecol Obstet 1997; 56:135-9.
103. Bennett KA, Butt KD, Crane JMG, Hutchens D, Young DC. A masked randomised
comparison of oral and vaginal misoprostol for labour induction. Obstet Gynecol
1998; 92:481-6.
104. Adair CD, Weeks]W, Barrilleaux S, Edwards M, Burlison K, Lewis DF. Oral or
vaginal misoprostol administration for induction of labor: A randomized trial. Obstet
Gynecol1998; 92:810-3.
105. Wing DA, Ham D, Paul RH. A comparison of orally administered misoprostol with
vaginally administered misoprostol for cervical ripening and labor induction. AmJ
Obstet Gynecol1999; 180(5):1155-60.
93
106. Butt KD, Bennett K.A, Crane JM, Hutchens D, Young DC. Randomized comparison
of oral misoprostol and oxytocin for labor induction in term prelabor membrane
rupture. Obstet Gynecol1999; 94{6):994-9.
107. Curtis P, Evans S, Resnick}. Uterine hyperstimulation: The need for standard
terminology. J Reprod Med 1987; 32:91-5.
108. American College of Obstetricians and Gynecologists. Fetal distress and birth
asphyxia. ACOG Committee opinion no. 137. Washington, DC. American College
of Obstetricians and Gynecologists 1994.
109. Hodnett ED, Simmons-Tropea DA. The labour agentry scale: Psychometric
properties of an instrument measuring control during childbirth. Res N urs Health
1987; 10:301-10.
110. Zieman M, Fang SK, Benowitz NL, Banskter D, Darney PD. Absorption kinetics of
misoprostol with oral or vaginal administration. Obstet Gynecol1997; 90:88-9.
111. Danielsson KG, Marions L, Rodriguez A, Spur BW, Wong PYK, Bygdeman M.
Comparison between oral and vaginal administration of misoprostol on uterine
contractility. Obstet Gynecol1999; 93:275-80.
112. The American College of Obstetricians and Gynecologists. ACOG Committee
Opinion. No.228 Washington,DC. American College of Obstetricians and
Gynecologists
94
APPENDIX A: PEER-REVIEWED JOURNAL PUBLICATION
OBSTETRICS& NECOL8
Volume 92 October 1998 Number 4, Part 1
A Masked Randomized Comparison of Oral and Vaginal Administration of Misoprostol for Labor Induction
KELLY A. BENNEIT. MD. KIMBERLY BUIT. MD./OAN M.G. CRANE, MD, DONNA HUTCHENS, AND DAVID C. YOUNG. MD
O"j~cti11r. To tat the null hypodlnia llut Mlnlnialt!rifta miMtp--..1 onlly w w.pn.Dy will rnult In 110 dlthmtce in time to wqinai birtll, aad to cl.tenlllna whftlaK cll._t E-.~n of tadlysyMole aad hypentimulation Ul! -
cialecl with route of ~ti-. M~rltotlr. Two huaclnd 1ix women &Iter 37 c-pletecl
wnka' ... ~- who praentecl with an inclicatian for includion were raadolnly ..Upecl to receive ..u.apra.tol 150 "" .tiller orally or wqiMlly every 4 houn • anclecl to induce lAbor. Pbcctlo taM anci.Uocalioa cancul-t aa:o_. plilbccl bllnclln&lllltil clata .malJiia was completed. SUiple size was cakua.tccl to .Uuw a two-tailed a of .05 ancl power n - #I) of 10"1.. All fwtal heart nre &ad uterine actirily pph• were d-l&cd aa:ordin& to Curtis' criteria bcfon lncludion sr-pa- uneukcd.
Rnwlb: An.aly.ia inwolwecl liM w- in lh oral poup anclliiZ ia tile nput pnp. 1M -an time I~ IWiclald clcwiatianl to wapw lridh with oRl Misop,.... - 1072 (~59Jt Rliftulft cam.,..ct with M6 (~) Jain .... with the vapaal protocoliP • .GIN). TMre - ao 1ipi&allt dUfa·
From IM Dtpllrtmmt af Obsttlrla IUUI GIJIVallogy. Mtmorud Uni· t'nSif!l. St. fohn 's, Ntw{rnuuJlanJ. Gmaota.
Suwrmnl in pr1rf by 11 p•tl from T1w HGJith Carr Corporatrrm of St. foltn' s tmJ Almranrrl Unrwnrty af NtTD[ountlbrrri. F«sdtrJ af Mtdiant. St. fohn 's, Ntulfowrrdlintd Calllldll.
VOL 92.. NO.4, PART I. OCTOBER t9'HI
ftiCn in caAnan rate, epidural taM, or neonatal outeD.._ Mon frectucnt Udly.y.tolc frw 20 lllinuta IP < .011 and hypcntllllalalfon IP < .IMt were oftclwd wilh v.pQJ lllilopro.tot. No -tal asphyxia occulftclln ctthcr potap.
CoiiCiuior. Miloplmtol dectiwly Induces IUar, pwcn orally or wap...Uy. nae ~ • ahomr iatcrnl liD nsinal birth with v._maJ •pplkalfon; howewr, the - fnqucat ocaarnac:e of fetal hurt rare snpls .tla-.litin ill th~ paup IUillfltl th.at. until the opti..W cloliaa iatnwal for vapaaiuc !. clctftDiiaecl, the preferred l'll1ltl of llliloptoltDl Ml•iailtnlioa llli&ht be oraL IOINt.t CfiiiCOl 1991.-92: 481-6. c 1"' by 1M Aalakaa Callcae of Obstctridand CynccolopMI.I
Mi5oprostol (Cytotec; Se.ule. Oakvillt!. Ontario, Cana· da). a synthetic prostaglandin (l"G} E1 analogue cur· rendy used for treatment of gastric and duodenal ulcus, can initiate uterille contractions. It is mMketed in an oral formulation of 100. or 200-~&g tablets with a maximum recommended dose of 1600 ,.g per day.1
lnduction of Llbor by placing a portion of 5UCh a tablet in the vagina has been studied extensivelyl-1; however. few randomized tn.Js9.1° asse55ing oral misoprostol for
00l9-71Mf/98/SI9.0D 411 Pll ~71Mf(98)011226-9
95
induction of labor at term have been published. Our research group compared oral misoprostol to vaginal dinoprostonc (PCE2) and found oral misoprostol to be an effective induction agent with no important adverse maternal or neonatal outcomes ... The median medication cost per patient receiving misoprostol wil5 at least 100 times lower than the median medication cost per patient in controls receiving dinoprostone.119
Despite accumulating evidence of effectiveness and obvious cost advantagr. misoprostol is not yet approved for induction of labor at term in most North American health care centers or by a national agency. Some authors have reported that tachysytole11 might be a frequent occurrence when vaginal misoprostol is u~ to initiate uterine activity,5·o.1:.13 which has led to concerns about potential increased rates of operative delivery and neonatal asphyxia. Any new induction technique must address these Stlfety concerns.
Substantive maternal and neonatal primary out· comes. such as rates of ces.lrean birth and birth as· phyxia, are relevant when cvilluating any induction agent. To test such outcomes using misoprostol would require Stlmple sizes of 3400 and 5030, respectively, because of the infrequency of these events. Before conducting such a lnge and costly project, it is necesStlry to determine the most appropriate route of misoprostol administration. For this purpose, we designed a masked randomized trial with time to vaginal birth as the primary outcome measure.
Our primary research question wu whether there is a -l·hour difference in the interval to vaginal birth with oral compared with vaginal use of SO ,.g of misoprostol administered every -1 hours for induction of labor in women at term with intact membranes. We also determined whether the frequency of excessive uterine activity resulting in abnormal fetal heart rate (FHR) tracings is influenced by route of administration. Other secondary outcomes included neonatal morbidity (as measured by cord blood add-base analysis and ACOG criteria for birth uphyxia 1~). cesarean birth, maternal gastrointestinal side effects, and patient Stltisfaction.
Met/rods
The study was conducted at the Grace General Hospit;:d. St. John's, Newfoundland, Canada, from March 3. 1997 tu October 8, 1997. This referral centerfor perinatll care serves an almost entirely white population of 500,000 and is the site of 2500 of the province's 6000 annual births. Our induction rate is just over 20'ro, with approximatcly 520 inductions per year. The research propos.ll was approved by the Human Investigation Committee of the Faculty of Medicine, Memorial University of Newfoundland, and the hospital
411 leftnett et at Mi:scprostol for lAbor /rrdudwrr
Patient inclusion criteria were indication for induction, single live fetus, gestation longer than 37 completed weeks, intact membranes, and cephalic presentation. Patients who presented with any one of the following were ineligible: nonreassuring FHR tracing, previous uterine surgery, known hypersensitivity to prostaglandins, age less than 18 years. or contraindication to vaginal birth. All pregnant women who attended prenatal classes were informed of the study and given written information about the protocol. Eligible patients were informed when induction of labor was indicated. At presentation for induction, patients were enrolled in the ~tudy by the attending physician, who explained the protocol and potential risks and benefits before obtaining written consent.
Opaque envelopes containing a card indicating group allocation were prepared by an administrative staff member using computer-generated random number tables with randomization in blocks of four. Allocation of each consenting patient to induction method was determined at the time of induction by opening the next sequentially numbered envelope. Study medications and placebos were prepared by a pharmacy technician. For each patient, eight powdered so.,.g doses of misoprostol were placed individually in paper packets and stored in an airtight plastic container. Eight doses of an indistinguishable powdered cellulose placebo were similarly prepared and separately packaged for each patient. Oral or vaginal use was specified on the label.
To conceal .tppearance and taste, all oral powders were mixed with 30 mL of apple or orange juice according to patient preference. Vaginal powders weN suspended in hydroxyethyl gel just before use and delivered to the posterior fornix using a vaginal applicator. After each delivery, the study package was returned to the pharmacy technician to ensure each participant had received the assigned treatment When a subject required more than eight doses, another identic;llly prepared medication and placebo package was provided. To maintain allocation concealment, research personnel who prepared the envelopes and medications were not involved in any other aspect of the study. The induction method was concealed from .til caregivers, participants, iUid investigaturs until data analysis wu completed.
After enrollment, subjects were examined to determine Bishop score15 and sb:ttified to either the low (less than 7) or high (7 or more) Bishop score group. Each patient was then randomly il55igned to receive eithu oral misoprostol (50 l'g) and vaginal plicebo or vaginal misoprostol (50 ,.g) and oral placebo every 4 hours until the occurrence of one of the foUowing: a contraction frequency of three per 10 minutes, a nonreassuring FHR tracing, spontaneous rupture of membrane, or deliv-
96
ery. All study inductions were conducted immediately after randomization. on an inpatient basis, to allow continuous electronic FHR and uterine contraction monitoring for at least 1 hour after each dose.
Decisions regarding amniotomy. analgesia. epidural anesthesia, and oxytocin augmentation were made by the attending physician. To assess patient satisfaction, all patients enrolled in the study were askLod to complete a modified labor agentry scale questionnaire.'~ Planned treatment of tachysystole and hypetstimulation included change in maternal position to the left lateral decubitus. oxygen 01dministration by nasal prongs. and intravenous ritodrine, as needed. Fetal scalp blood sampling was performed when indicated by non reassuring FHR graphs. Cord blood samplt.os for arterial and venous acid-base analysis were taken after eao:h delivery.
To determine whether hypertonus. tachysystole. or hyp:rstimulation were a.~sociated with the route uf administration of misoprostol. all FHR graphs were reviewed before study induction groups wen: unmasked. Each of the physician authors (two maternal fetal ml!dicine attending physicians and two senior residents) independently classified monitoring strips using terminology defined by Curtis et al11 and Winget al" as follows: hypertonus. a contraction lasting more than 90 seconds or more than 120 seconds; tachysystole. a contraction frequency of more than fiw in a 10-minute period or two consecutive 10-minute periods; hyperstimulation, exaggerated uterine response with lah! FHR decelerations or fetal tachycardia greater than 160 beats per minute. The code for group olS.~ignment was not broken until completion of data analysis.
For the primary outcome. time to vaginal birth. a sample size of 172 was calculated tilling a two-tailed a .. .OS. 13 • .20, ;1 = 2-10 minutes. and u derived from pooling of results from our" vaginal misoprostol cohort (588 minutes) and our previous oral misoprostoloJ intact membrane stratum (524 minute!~) (PEPI, Version 2. 1995; Computer Programs for Epidemiologic Analysis. Stone Mountain, GA). Since our previous studiesu had consistent cesarean rates of 12.2'ro with 99~o confidence interval (C[) 8.8, 16.5o/o we added 20'Yo (3-& patients) to aUow for anticipated cesarean births in our 5amplc. A sample size of 206 was deemed appropriate to detect the clinically important ditrenma: (2-10 minutes) deter· mined by a survey questionnaire of patients and medical and nursing personnel.
Dilta were ilnillyzed on an intent-to-treat basis by parametric and nonparametric statistic;, using Statistix .U (Analytical Software. Tallahassee. FL). Ol!dsion levels and hypothese to be tested were preset to minimize bias. Hypothesis testing was perfurmed on the primary out· come. Other comparisons were considered hypothesis
VOL. 'f2. NO . .f. l'ART I. OCT06ER 1993
Table 1. Demographic ~racteristics
V.gin.a! misoproslol
Owxmistic (II • 102)
Nullip•""'-" M.tlrm.ll ~so: (y)
CeltAtion (d)
Gr.avidity PJnty Bishop ocore lea th.ln 7 lndiclnon tor induction
Post-term Hypermt.""'n Oligohydr.amnom Oth«
Birth w~ght lgl
7-6 :!11.7("-9)
2llol.l (8.3) 1.6(0.11) 0.3(0.6)
81
6S I! 8
It J6o&5 (566)
D;ara Rlvm a numbrr or me•n (stuldud ctev;..Jion).
cma misoprosiOI (II • 104)
69 27.5 (5.0)
:!a$.4(7.6) L7(1.0)
0.5(0.8) ~
69 12 13 10
l585(6Jl)
generating. Statistical significance of the primary outcome measure was assessed using Student's t·test and was considered significant at P < .05. Rank order non parametric statistics using the Mann-Whitney U test, when a Cl!Silrean birth is a failure to deliver vaginally and ranked longer than any vaginal birth. aUowed inclusion of an births in a !iC.'CQndary ilnillysis with median time to vaginal birth as the measure of centralllmdency. Secondary out· comes were analyzed by parametric and nonparametric stltistics. as .1ppropriate.
Continuous variables were examined for normal distribution (Wilk·Shapiro/Rankit Plot) before using parantetric statistics.17 Desc:riptive statistics were used for demographic baseline data. Baseline data from nonvolunteers were analyzed to assess generalizabUity. The significance for aU secondary and hypothesisgenerating analy5e5 was P < .001 to account for multi· pie testing. a conservative approach.
T.able 2. l'eripartum Oatil
95'!1. V~gin.ll Or.ll Rd.lliv~ CDIIfodencr
Vuubk! (II • JQl) (rr • 104) nslt inlerv~
lntnp.artum fmtumcy (h~;n ... S9 i'tJ 0.67 0.38. J.IJI Epidural usc 61 6S 0.89 051. 1.56 No ouoalgew II3Cd 6 10 1).59 0.21. 1.64 :1.1.-cuniwn 21 29 0.78 0 .• 1. 1.47
P•onlWilltn~
Ef!i.'iotomy Z! 19 l.lJ 0.62..:t41 l..acn~ S4 S9 11.98 0.56. 1.69 Third· or fourth· l 0
dqtft l«uallon Intact pmrcum l4 29 l-29 0.12.l.ll
lanett et II MisoprostDl for Uzbor lrrducticn 4IJ
97
Tible 3. l..Jbor !ntervm to V;~ginal Birth
Induction to votpn:al birth lnductlan ro full diLation First s~p: of LJbor 5«ond stage of Llbor Third s~c of t..bor
V•gmal misoprustol
846(3851 :'63(361) 3-l2 <:W..I
8.1 (119) 12(1-1)
10'72(593) 1003 (3711) 365(223)
69(76) 9 .q(8.-l)
0.~ .anr givt:n in m111utes .u rnc:an (sund.vd dni.lriunl. • ~lcuY~ by Student t rorst.
Results
.001
.002
. .as
.:!6 21
During the study period, 393 patients presented for induction of labor, of whom 308 were eligible. Seventeen patients refused enroUrrumt because they did not want to be involved in 01 research protocol and the remaining 85 were not invited to participate because altem:~te methods of induction, including ilmniotomy, oxytocin, or dinoprostone, were preferred by their attending physicians.
Of 206 women, there were 104 in the oral group and 102 in the vaginal group. No participo:~nt or neonate wo:~S lost to follow-up. The mean time (!:standard deviation) to vaginal birth with oral misoprosto1 was 1072 {!:593) minutes compared with Sl6 (!:385) minutes with the vaginal protocol. a statistically significant diJFerence (P = .004). ("The median time to vaginal birth in the oral misoprostol group was 1125 minutes compared with 962 minutes in the vo:~ginal misoprostol group [P = .38, Mann-Whitney U testJ.) Maternal preinduction and neonatal demographic data are given in Table 1. Periparium data are given in Table :!. with relative mk (RR) and 95'Yo CI. There were no significo:~nt dilferences in
T1ble 4. Cli155ifiation of ExCI!Ssive Uterine Activity
Tracing dua&f&e.ailun
Hypertoni.: >'ill S«
Oral misup~rol V.agin.ol nusopmsrol
H)l"'fllNC >120""" Oral mixlpruMOI v .. pna~~
T.adlysystolc > 10 min Oral milop-tol ValiN! nusoprD11tul
T.adlysystolc >20 min ()r;a! milopi"Dl'llDl Vaginal 11\UoprastDI
Hypcn!Unulallon ()r;a! mi5uptollol Vq~NI mi5opt"oltol
7 8
23 16
52 41
M 70
Rnlew~
cl.ullifird graph a
8 12
~
21
1-1 15
21 17
1-C 17
Table 5. Neonatal Outcome
ACOC criteria for birth .asphyxia
Appr .at 5 min :sJ Coni pH <7 IJ.uoe ddicit >16
Mean CURl pH Mr.an b;uor dclicit Mectw. Appr .11 1
minute
V.agiN! misoprustol
0
7.28(0.10) 5.2JI(U4)
q(8.9)
a 0 0
7.30 (0.09) U4(3.J2)
'1(8,9)
p
.16'
.16'
.08'
.ll' 59'
q (9. lal 'l(q,101 ~ Apg;u .11 5 .231
minurc Apew .11 1 nun <7 17 mt Appr .11 5 min <7 a .32'
• F".stwr ...w:t ret • Studl!flt r ret: """"" (st;an.Urd dev~tioa). 0 1\.t..M·Whitno:y U ret: mt<dim (lint and !Nrd qu.anile). ' .r' lftt.
oxytocin, epidural. or other analgesiil use. Data for mean vaginal birth intervals are given in Table 3.
There was no significant difference in birth route between the two groups (X,1 = 2.53, P = .-17). With oral misoprostol there were 58 spontaneous vaginal deliveries, 21 vacuum deliveries, nine forceps-assisted deliveries, and 16 cesarean births. With vaginal misoprostol there were 56 :;pontaneous vaginal deliveries, 15 vacuum deliveries, eight forceps-assisted deliveries. and 23 cesarean births. Nonreassuring FHR tracing was the indication for two cesareans in the oral misoprostol group and six cesareans in the vaginal misoprostol group (RR 2.-17, 95.,-o a 0.-19. 12.49).
Fetal heart rate and uterine Ktivity tracing data were
.. Reviewft!' RcvicWt'I'J illr'v...,en cl.aulfied d.IA.•II'Wd dalsi!Md gr;aph.as sraphu graph .as po
15 :!6 5& .611 II 2IJ "" :!-I ::!0 31 S1 ~ ~ 21
IS 25 :0 .21 Ill 21 31
16 10 5 .a1 7 11 :!5
0 .Ool 6 ..
analyzed using nonparametric: statistic:s based on number of masked physician reviewers making that classification of a given tracing (Table -1). Although there was more frequent 20-minute tac:hysystole {P < .01) and hyperstimulation (P < .04) in the vaginal misoprostol group. these results were not statistically significant by our preset level for secon~ry analyses. but the trend warrants attention. Neonatal outc:omes in both groups were similar (Table 5). Umbilic:al cord arterial and venous blood acid-base analysis was performed on 89 of the 104 participants in the oral misoprosto1 group (86% of neonates) and 92 of the 102 participants in the vaginal misoprostol group (90'Yo of neonates). Two neonates from the vaginal misoprostol group had c:ord pH less than 7.00; however, no neonate met the ACOC criteria for birth asphyxia.'•
A postpartum questionnaire was c:ompleted by 90 women in the oral group and S3 in the vaginal group (84'Yo response rate). Analysis of the responses to the -49 questions failed to deb!ct any statistically !!ignificant dift.'f1!1\Ct! in patient satisfaction or overall score. All participants in the study reported positive labor and delivery experiences. None of the p;uticipants self-reported adverse gilS
trointestinal elects on the questionnaire. There was no dianhea nctl!d in either group, and no diference in ~"CUffee''IZof vomiting (18 in the oral and 19 in the vaginal group. P = .81). When given orally. a nwdian of two and a maximum of nine doses were used. Only seven subjects took more than five oral doses. When misoprostol was placed vagiNlly, a median of two and a maximum of five doses were used. Bl5eline data from nonvolunteers did not difer significantly from volunteers.
Discussion Obstmicians routinely use various mechanical methods and pharmacologic agents to induce labor; however, no single approach has been universally successful ()l(ytoc:in is an ektive induction agent but it is not optimal for patients with an unripe cervix. uu9 Administration require intr.aveNNS infusion with continuous monitoring of FHR and uterine cattr.Ktions. When campan!d to oxytocin. prostagLandins (dinoprostone) are e6«tive cervial ripening agents, 1c:hieving a shorter inductiondelivery intftVal and a lower rate of operative intervention. :zo Oral admini5tration of prostaglandins ~11tiaUy has been ab.ndoned because of undesirable gastrointestinal d'ects, which have limited their use to vaginal and intracervical routes. 1' Tennination of prostaglandinrelatll!d excesaive uterine activity remains problem.ltic. :n
An oral induction agent that is safe. e~ective. inexpensive. and weD-tolerated by patients would be attrac· tive to patients and health care providers. Anticipated benefits include avoidance of intravenous lines in some
VOL 9Z. NO. f. PART l. OCTOIIEK 19¥8
parturients, less frequent need for vaginal examination, and greah!r freedom for upright positioning. Ambulation might even facilitate labor progress.::z Induction agents shouJd mimic spontaneous labor while avoiding l!xc:essive uterine activity. When c:ontractions of abnor· maUy high intensity or frequency oc:c:ur, insufficient recovery resulting in FHR abnormalities could necessitate operative delivery or lead to fetal asphyxia.11
Several r .. ndomiud trials2"' comparing vaginal misoprostol to standard therapy found misoprostol to be safe and c!«tive for induction of labor. Sanchez-Ramos and colleagues'~ studied the safety and effiacy of vaginal misoprostol for cervial ripening a:ld labor induction in a meta-analysis of published trials to 1997. Of 16 trials identified, eight met their criteria for inclusion and included a total of 966 patients. 488 of whom received vaginal misoprostol for labor induction. Controls received oxytocin or vaginal dinoprostone. Those who received vaginal misoprostol had 3 higher incidence of vaginal delivery within 24 hours of application OR 2.6-1, 95~ .. Cl 1.87. 3.71) and a lower cesarean rate (OR 0.67, 95'l'o Cl 0.48, 0.93). Use of vaginal misoprostol was associated with a highe1' incidence of tachysystole (OR 2.70, 950ft. CII.SO. 4.04) but not ofhyperstimulation (OR 1.91, 95% Cl 0.98, 3.73).
Only a few randomi:~:ed trials assessing oral m~ prostol for induction of labor at term have been publisht!d.9•10 Toppozada et al13 compared vaginal and oral misoprostol for induction of labor in a smo:U randomized trial. Twenty patients assigned to receive 100 I'S of vaginal misoprostol were compared to 20 assigned to oral misoprostol. When no response was noted 3 hours after the first lQO.I'K dose, 200 l'g was administered every 3 hours until a maximum dose of 1000 l'g was reached. Doubling the dose administered .. -aginally only occurred once. One subject in the vaginal group and three in the oral group had cesarean deliveries bec:ause of failed induction. The authors concluded that vaginal misoprostol resulted in 3 shorter time to delivery (P < .005), but more abnormal FHR patterns and uterine hyperstimulation occurred (P < .05).
Zieman et al23 r.ompared pharmac:okinetics of misoprostol (400 ~£8) administered by vaginal and oral routes to ten nonpregnant and ten pregnant ftrsttrimester volunteers and conc:luded that significant differences exist. Systemic bioavailability of vaginally administered misoprostol is three times that of misoprO!Itol administered orally. The greater bioavailability of vaginal misoprostol might explain why we observed a shorter time to vaginal birth in the vaginal group. The greater frequency of abnormal FHR patterns observed in this group might be the result of excessive uterine activity because of this greater bioavailability.
Although there is no direct one-to-one relationship
99
between continuous electronic fetal monitoring and fetal well being. nonreassuring FHR changes associated with excessive uterine activity warrant intervention. either to l'l!dua! uterine activity or to deliver the fetus. BecaWJe of confusion from varied definitioN of excessive uterine activity, Curtis et iil11 recommended a standard ll!nninol· ogy. Although it allows for classification of labor graphs, the terminology is arbitrary, particularly with respect to tachysystole and hypertonic rontriM:tiol1s.
Despite concerns of po551ble uterine hyperstimulation. substantive advene newborn outcomes such as neonatal acidosis, meconium aspiration, and birth asphyxia have been rare and have not difered between misoprostol and control grt111ps.•-•o.tl..l3 1n our study, two neonates in the vagirlal misoprostol group had cord blood pH less than 7.00, but no neonatal asphyxia occurred. Among the 39 cearean births, six of eight done for FHR abnonnalities were in the vaginal group. Although these difeteN:eS are not statistically significant, the quetion is clinically relevant, making it worthwhile to design a clinical trial to answu this question. 1l1e fn!quency of exce55ive uterine activity ISIOCiated with route of administration is a sec· ondary outcome in the present study and did not fonn a basis for cakulation of sample size.
Misoprostol is elective in inducing labor whether it is given orally or vaginally. There is a shorter interval to vaginal birth with Vilginal application; however, the more frequent FHR graph abnormalities in this group might be attributable to excessive uterine activity. Oral administration is an appropriate alll!miltive for labor induction with misoprostol. considering fetal safety and the ongoing study of optimal vaginal dosing interVal.
References
1. c.ms RE. IC1rkwood CF. Mimproslo~ A proatag1Jndin E, ~Josuor. Ctin Ph.lnn 19119;1:627-.W.
2. Sandwz·IWnos L. IC.\unitz AM. Oo!l V.Jt.. GO, Oo!lb I. Sdlrocdt'f PA.IIrioNs OK. r....bar induction wath !he proetagWndin E1 methyl an.~los- miwp.,.IIJI vorn ... oaylOCin; A ~ trW. Qb,tet Cyn«oll993;111:332-6.
3. l'ldchft HM. Mitdwll S. Simo:on D. Fm:lmck I. Brown D. Intra• vap!Al miJcprmlol • • <ftVial ripcrung •genr. Br I otlstet C)'N«<l 1993;100:641-...
.a. Ft.uher H. Mirdwll S. Fn!do<rick 1. Simeun D. Brown D. lntravag·
iNI ~· vas ... dillaprmrarw u cwvial nPfttins and IMior-induc:illc .aca\ls. 0bsC1!t Cynecoll994;113:24.f-7.
5. Wlllg DA. lonn MM. RaiWt A. Coodwtn TM. Paul RH. A compu!MX' of ~toland proerapndift E. pi ~ pmnduclion ...mal npmifts and Ltbor induction. All\ I a...t CYJWCOI 1995:17l:IIIIM-10.
6. W"UtJ OA. IWYII A. IONS ~L't. Coudwin n.l Paul RH. MisoplDiltol: An decti.-. "'Fftt for CIUVial ripenin& and Llbor induction. Am I Chtl!t Cynecol1995;172:1811-ft.
1. Vuaklls K. CumiN R. Srubbldirld PC. Randomizltd controlled rri.d of vapw ml5opta11DI mot inlr.JcftVical prosrasJ.anciin E. ~I for induction of Ltbor .at r.mn. Otlsr.t Gynccol1995;86:541-t.
8. Mond~ WR. Y"liiiW DC. VolliNI rnisopftlllol for inductim ni Llbar. A ~omiad cmtroUed trW. Obllzt Cyrwcoii996J8;521-5.
9. Windrim R. l!cnnett K. Mundle W. Young DC. Oral ~dministratiun of mlsoprosiDI for Labor induction: A randomized controllcd tn.L Obsll:t Cyn«al 1997;119-.392-7.
10. Npi SW. To WK. lM> T. Ho PK. CerviGII priming with om mwopra.ral II\ p~L1boT ruptun. of ~Mtnb.,_ ~~ twm. Qb,rer Cyno:ol 19'16;87~..3-6.
II . Curti~ P. Ev•ns 5. Rernck 1. Utmne hypcntunuLnion: Tho: no.~ for ~bncbnt rc:munology. I Re-prod Mord 1987;32:91-5.
12. s.nchft·Rmwa L. !Yurlltz A. W..aD R. Drlk I. C.udioer F. Misoprostol ~ =vial ripcroins •nd bbor inducticm: A m.vWym. Qb,lft Cyn«ol 1997;89:6»-·U.
13. Tnpp<IZii<U MJC. Anwar MY, H.asun HA. El~rly WS. Or.ll llr vagtn.ll mioopnbii:JI for induction of t.bor. lnr I CynMCOI Obtcrt 1997;56:1JS..9.
1-1. AmaiGin Co~ of Obstdr~CW>s and Gyn«ooopto. FCIOI1 disand bum .uphy.a.. ACCX: aJINilire... upinioxt nu. tl7. w~ OC: Amman c~ ot ~ .wf Cyn«ologisto. 1994.
15. Banop EH. Pelvis SC'Orins fnr elective inductinn. Obsrer Cyn«ol 1%&;24:266-8.
16. Hodnarr ED. Simmons-T~a OA. The bbour •gmrry w:;ole ~ychumctric propmia of .on iniUW11Cnl lllftSunn~ amrrol durinS childbirth. Re Nurs H~r.Jrh 19117;111'.301-10.
17. SlundeR BD. Tr~pp RG. Par:unrtric and ncr~p.~ramc>tnc: >t.Jftlrio. In; DoLin I. !..utpn C. •-.b. !laic .and Clinicill Biost.on:ltics. :!net rd. Norwalk. Conrwcticul: Appleton .. wse. 1994:99-12-1.
18. ICeirw MINC. CNim..rs L Mdhods of inducing ~bnur. In; Ch.Jim· ~rs I. Enkln !'wl. tc.i:w MINC. ords. ~ve Gin! in Pft~JW!'Y •nd duldbuth. OUotd. Unill!d Kingdom: ~ford University Pras. 1'111~1057-19.
19. ICeirw MJNC. v..n Oppom ACC.I'rwpUUtg tlw atn~ll< fDr anduction of ~- In: ClWrnftS I. Enkln M. kleilw MfNC. ~· Efecn•·e cant an p~ .uut childbirth. Oaford. Unill!d Kingdom: 0..· lord University P-.. 19119:988-1056.
:!D. ICeirw MJNC. PrasugLandirul in pmnducnon C\!f\'tal ripenins: M.r:t....W)'U of worldwtde dlnical l!llpommce. I R.prod Mord 1993;38 Supp~fi-911.
21. Eprrer CH. H..-t.in PW. Raybum WF. Ub!nrM! 1\Yl"'r.ollnud .. lllln atb!r low dc.e pmsasl.andin E. dwnpy: Tucclylic lre2tmmt
in 181 a-. Am I Obstet Cyrwcol 19'10:163:19-1-6. 2:!. Sfiby H. Upripr venus ....:um~~mr position durins dw first~"'~
of Labor. :n: Enkln MW, ICeirw MJ, Rftllnw Mf, Neilson IP. ftb. Pn!8NftCY .uut childbirth mud..W. Oxford. Unill!d Kingdom: Codlr.alw 0.~ uf S~bc: Review~ 11!VII!W no. WJ:W. 1'i'J.f.
ll. Zieman M. Fang SIC. Bomow1tz !lo"L. B.lnsk~ D. D.uno:y PO. Absorpdon klnetia of mi»uprosii:JI wtth oral or ··~pyi .lldmin~Strallon. ()b,tef Cynca~l 1997.90:311-on.
Addl'l!55 reprint requests to: klly A. Bmndt. MD Dqartmort uf Obstmrcs md Gyntcalugy Ylillt Uni~ity Sdrool of lv'~icint J33 Ctdllr St.. PO Box 208063 Ntw H~Wm. CT 06520.8063
1Uaiml MArch 10. 1998. R«riMM in rm-1 frmrr ,O.Liy f. 1998. "'aptet ,o.r.ry n. 19!18.
Cupyrishl 0 1998 by n.... AJncric:an CoU~ of QbsrerricW!Is and Cyrwcolopts. Publishm by E!MvWr 5cicnce Inc.
Obsldrics & C~
100
APPENDIX B: HUMAN INVESTIGATION COMMITTEE APPLICATION FORM
MDIOIUAL lJNIVDSII'Y OF NEWIOlJNDL\ND • PACIJLTY 01' MmiCINE
HJIMAN INYf.STIGA'DON COMMrJTEE - c\PPUCATION FORM
11lis fimD is delia-1 10 cover • lqe a vlriely of propollb • poaible: DOt all quCitioaa ~L-.~Ij.~i= bowever, please coalider CICb qaeatioa cuefidly befOre writiDc it off • •Not Applic:able•
l. Name of Priacipllllaveaipor: ----..al.u«lbru...;:t .. ,_, __ L----------Mailill1 Addraatrel...,_ No.: 13 un 1qy Rpqtl a , •. , Nl ang.J221
2. Name(s) of Co-1aveati,.0r(s): .~P.:~'-...acDqyl4aall.llr-~~~~...,, -------------
3. N..,. of supervisor, if Priftcipallavestiptor is a studall: Dr PaWl YOMIJC
4. Title of investiption: (PI I&SJ HHJGIII' 1JY WI!IQS)
Vsrtnql Hl.rwmpql Vmw Orpl HLrtpwtqt For lrttl¥Q1911 Q(IMqer· A Rqadqm4td Owrqi/M Tripi
S. Whal is die proposed startial dale? (Must be iL.Ja1S 4 weeb larer dian !bee of receipt of tbis applia~~ion by tbe H.I.C. Office.)
'!qpgry ,.
6. Whal is lbe amiciplled dale of oompleion of dae study? 1411HQ1Y 1227
7. Please fill in dae lfProprille informallon:
a.,~~a~~c-•., ...... laftlftl ...... s.lllalu.llo Parddpadna Sealaa laftlftlll ......, ._. Flellllls lbpltal Eabia Commlu.a ........ Grice Geaeral Hosp. Yu r~.r Yu Yu
Labour a: Delivery
Uait
101
Pqe2
J7w ,.,., obJ«<iw of tills llll/ly II ID CClfi9JCft dw t/lk«J of lfllsDprom1l (CI pi'MIQJi4lltlbl) adlrtlltsluN DNil1 ro Mlsoprollol ~ 11111,_, lit ptllilllll f'Wllllrilll arvtctll rlpalltJ.ftJr rtw lltdut:ttolt ofhlbor. J7w tJII'1ID# 1110 tlnmlfllv 1/rtlfM of~,.,.., MJliOII dw time ofilttbM:Itoft ID dlllwry. 1111 ow COfWIIdoll dtlll e111 orGl lflltllcodofl Wlllfll4 be ~ to MIIIWII tu II coflld tl«ntu~ dw ,.,..,. of W111M1 ,.,., 1/wy """" raJflft llrnlfllll * llttl¥altJfa prtJUD.
9. Wlllt is die scitlllifk: bal:qrouiMI..a ndoalle fDr tbeiiiUIIJ? Wlllt beaeftll may be IDliclpated'P Have aay reiiVIIII tmm. or aaiJIIII lllldielllrady bela """"'"""" Pleae specify. (Aaadl aaodlcr s1aeet of paper, if required.)
P'ro.rtllrl4alllu tW boMI ro be qftalw bttlwrltJia .,a. Slwralltllllln haw shmm 11111J1U IIWoprostol ro be "* Gild~ lit anta~~ f"'palltrtlllll ,._, btdMt:rlott lit ptllkra """ illltiCIIIItJ!Iblau, iltdudbtr 011T mar's RCT. Mlsoprostollltlbofltl DIW pntnl tu costly tu Ollwr GfalS.
A """' RCT compi«ld ell tlw CiNe~ Gtltmll Hosp#IQl hCI.r shoMI ortllllflsoprostol ro be 1111 ejfocttvt! l1tdualoll CIJtltl. No rtUfdolrtlud lrillls llcM Cl1lllptUftl ONl lldlrtbtLrlrellloll to 1111111N11 adlnlltlstratlo11. 77W study GllllllpU ID dJu1fy Wdwr 111111111C1l or ONl GdlllllllstrGif II 1M pnftmd rOIIU of #UIIwry.
10. Wbicb of die followiq are 10 be employ.a iJt tbe iavesdplion? List only tbose tbat are NOT part of normal pllieal care.
(a) S1111pa. 10 be taken from subjecu: Sraae type of sample, frequau:y 11111 IIIIOWil.
No
(b) List die proc:edura IIIII IDY t1111 or subiCIDCa co be admiJlislend ro plliems: spec:ill dieu, dnap (stae dose IIIII fnlquacy), isoeo~Jic tncen, «e.
Grotip I.
~II.
Mtsopronol 25 111 PO ~ry two ltolln IIIIIU loboflr ~stllbllsMd. spoNIUWDIIS ruprur~ of trWIIfbrtuws ordlllwry.
MtsoproSIOI 25 111 PV Cr~ery 1W0 ltotln IIIIIU loboflr tsl4bllsMd, SROM or tUilvt!ry.
(c) Questioaaaira: Aaadl copy of qucstioDDaire 10 be used.
Ml«<ttd.
(d) Is tbia IPPiicatioD for a cliftical trial? ( K) Yes ( ) No If ya, wbll •p~~aae• of dle trial does tbis saady repraenc? Wha is tbe delip of tbe trial (e.g. opea. double blind, ell:.)?
Opm. l¥tlw J. ~ CDiflroiiTial
ll. Doa dle study involve tbe use of my radioactive llllhrial? If ya, specify.
( ) Yes ( .r) No
102
Pap 3
12. Give a brilfcr.cripdoDofdaedaip ofdae llUdy. (PI- abo lllldl ma.copy of a prococol if available.} 11lis sbould iad1lde c111aiJa of subject selection. .-pie size c:alculllioa (if appliclllle), OUCQHDe measuraaeat IIIII clllaiJa of lllllysil.
l7lt llttlallfiiiD'.r .,_, 10 COitlbla 41 1'GNlolrUud COIIII'Oiltd trl4l ctJ1f111Grl111 orQ/ tUtti WIIIMI mlsoprostol as bttlaalorl ~. /11 orrlu 10 e1mJ11111 for dw ~ of PGrllY Oft lltdMaloll of,., ~ wUI 1¥ .rnwiJI«l. 1\lo f'OIIPI will - ntllbll.rlttd, fllllllptJtuar ptlllnD lllfllllflllllptlnHu potlal.r. ~ 10 JIDII tlltd ...,.lfllloptwlol WDIII4 '- dtJtwiiJ "'*" .,.,..,U, ,.,.,rei optlqlltt mwlopa wllh Rudy tlll«tlllotl. l7ltu WOfll4 - pwptrttliiJ ~ .rlll/1 .. lllwJiwrllla pGIIal Clll'e MJiltl ,..,.,. ......,.14bla """ ~ Ill biDd:J of /Dflr. GnHip 41111,.,.,., Wllfll4 '- ct1ltt:4tlltd /fotlt Clllf! rtwn lllftU tlw lilll4 of ,.,.,.,. All Rudy p4IMitl.r will - odlrtllt«lto /wplltll tllflll1ttiMerlole will - Cllrrlal 0111 011 all llfpollal basi.r wllh ~ frtiiii'IIOIIllortJII for CNW 1tow tl/t6 ~ of t1w pronar141tdlll tUtti dwtltr atllblulwd lllbotr.
13. Number of sultjecu: Sr.e bow subjects will be selected.
780 lllbj«t.r wUI br rKn~Jttd j'rt1lfl ~ p~ tllfll obnmlcQ/ praatc~.r ill St. Jolul's. Pa*nn will '- ~llflbl~ lfdwy pn.r~llt wllh a11 btdictllloltfor lltdiiCI#oft of labor, a sillrl~fonu, c~pMJic pnsnuattollllltllllltaa IIWJIIbrG~VS. 'lPien """' br 110 COIIITtJbtdllaloiiiO WIIIMI blnh.
14. Number of coaaola: State bow dlcy will be seJcc:ted.
Euhuloll crilutG will I¥: lfDIIIYIWIITilllfntJI /won trru:1111, prior llleriM surr~ry. dot:JIIfwllud ~n~111ilivity to llll.ropronol or otlwr prosiGfltUtdLIU or COIIITGittdlcad011to WIILMI blnh.
15. W1la (a) risks, (b) discomfons or (c) inconvcnicnca are iawlved?
(1) Po.r.rll* ra.rti'OUWitilttll.rlde qJ'«u assodtllttl will& pronGrltutdbt illcliiM: t~GUSN, vom~~tnr. dlllrrMt~. (2J A rrn COIIIpllaJtlofl of pro#llfl4ltdlla use 1.1 •rllw ~ A MedlCQI Protocol for lllldres.rblr this slnMJdolt I.J uttlbli.rlttd.
16. Are lbere uy imlllediale beaefiu atiliq out of !be audy for the subjects? (Specify)
FrMT pelvic ~ Mirltl ullilll4ldy fWd to br pnfonrwd. Data from a Sillily diNw Gl our lllsLtllllo11 ~a ntlllnktll tUtti dllalallly slpljfcallt mhlaiDft of owr tJwe ltolln (p=0.0/8) ill tiiM frotra Utdw:rio11 to variNII blnla wiiiJ WJflltGIIffUDptollol tlll1rtbllsurtd 50 "' every four hofln PV.
17. What sups will be takca to preserve coafideatiality?
l7lt llnlatlftiiOn II.JIId wUl br tlw Dilly peopl~ to llluttiU dtllru tllld lwalth rrcord.r. No pGII~IIl /MIIIiflilll ~ wUI br rrletiMI for publlCIIIIoll or P'f#III4IIOII.
ll. Espa.ia pmc:eduae em obtaiaiq c:oDICDt.
Wbo will make !be iaitial co.act wida liNt subject? Alrgdlnr Plmlr:fa
Who will oblaiD die COIIICDI of die subject? '"""''flfllrJ 11m4 Ml[nn ml4ptplrclcm twac (If ftllldi1al .ruppon obtabttd)
103
19. wm lllbjectl illdude ...,.
leplly incoalpltlat periOIII
If 10, ........ will be raba to probiCt daeir riafds?
21. (I) Will ¥Oiullleal receive ~ for apenMS
time 1o1t from wort
Pap4
)Yes (.I')No
)Y• (.I')No
) Yea ( .1') No
)Yea (.I')No
) Yes ( .1') No
) Yes (X) No
•• Pleae speeify oa separa IbM Kemdina to GuideUDea for die Remuneraion of Research Subjects.
(b) Will dlere be 1DJ dlird pany remuaeraion for referral of pllieacs? ( ) Yea (X) No
•• Pl ... ll*ify oa scp ... sb• accordina ro Guidelillll for Payment ofFiDders' Fees.
•• AVAILABLE IN 'DIE OmCE OF RESEAaCH • GRAOOATE S'llJDID (MEDICINE)
22. PI_. enclose 1 copy of die buda• for dlil mady, iadudina an iDdicaion of source of fuadina.
Will die budpt be .tministered dlroup dae UDivcnity Fiaance Office? ( ) Yea (X) No If no,ll*lfy. 1/mNI'dl/flltdlllr obftlbtalll t4IOII/d 1» ~,llwllrll 1M Ulllwntly.
WUI die iJmltiplor accrue my beDefiU by virtue of plfticipalion in tbis saady? ) Yea (X) No
23. Is tills pan of 1 aaulti-cenlre study? )Yes (X)No
24. Will d• bec:o.- dle exdusive propeny of 1 pbiiiiiiCeUtical company or otber ouaide apacy?lf yes, please elaborate. ( ) Yes ( X ) No
25. It is tbe relpOIIS1bUily of die u.v.dpcor to IIIIUre Ill• permission is obcained from dillicians, depanments, illltiUioaa or COIDIIIIUiitiel wbose plliellcs/reaidenll will be involved in tbe study. Have tbe app1011rialc colllai:U .,_ llllde?
Yn
104
Office ol Raarda and Graduate Sclldia tMcdicillel Faculty of Mtdicine The Health Scialca Centre
1997 Olll KEYED
Dr. Kelly Beaneu Deputment or Obstetric:s/Gynec:ology Grace General Hospilll
Dar Dr. Bennett:
Thank ,au for llldna tbe time to complcce the annual update form ror 1he raan:h study entitled •v..-. Mllopnl1al Vtm11 Onl Mileprostal for laductioa or Lahor: A Ra•do..a.d Coatroled Trial".
At a meetiftl held on Pebna1'7 17, ltt7, the Human lnvestiprion Committee granted approval oftllia study Ulllil febnw'J 1991 at which time yau wiD be ccmlaCicd for a fUrther update.
Human lmaliption Committee
HBY\jalo I
c:c: Dr. K.M. W. Keou&h, Vice-Piesidenl (Jlacarch)
Sc. rc*a•L Nf'. Cmada AlllV6 • Td. : 17091 737-6762 • Fu: 110!11 7l7·50ll • auol: rpeiiiOfpA.IICI.mua.ca
105
APPENDIX C: DATA COLLECTION FORMS
Group
Chart##
Ap
Gravida. Para
Gestation (clays)
Bishop - Position
- Ccmsiltence
- Efliccment
-Dilation
-Station
##Doses ofNarcodc
Epidural
PG##ofDosea
Ma.Miso-Vq.
Ms. Milo. 0n1
Oxytocin ## miaates
Indication for Induction
Type of Delivery: SVD Vacuum Forceps LSCS
Indication for OR: NR.T FTP None
Episiotomy: Nil Lat. Midline
Lacerations
Manual Remcml ofPiaccnta
Blood Loss: Normal or
Scalp pH
Side Efl'ects: Nausea 1/1 ..
Diarrhea
QUES110NN.AIRE SENT _YES NO
106
-R.O.M. (Time)
t.feconium: Yes No
1st Sla&e (mins.)
2nd St.qe (mins.)
Induction to deliwry (mills.)
Induction to fully (mins.) •
saaaem
Apprl
Appr5
Cord Pb
Cord B.E.
Gender
Weight
I Pelvic Exams
107
INITIAL BISHOP SCORE· PLEASE CIRCLE
Parwnelers
Cervical Position
cervtca1 COnsistency
Cervical Effacement ~)
cervtca1 Dilatation (em)
Station of Fetal Head
Total Score •
Time and date:
Assessment by:
1
2
3
4
5
e 7
a
Oxytocin:
Time Started:
Bishlp Score:
Dose:
Gravidity:
Parity:
Geslallonal Age:
Time of Delivery.
0
Posterior
Firm
0.30
0
·3
1 2
Mid Anterior
Medium Soft
40-50 80-70
1-2 3-4
·2 ·110
Dos!p! Time
3
>10
>5
+1
Date
108
APPENDIX D: POSTPARTUM MISOPROSTOL QUESTIONNAIRE
( ( Appllcanc's Name : Kelly Ansela Bennect
POST-PART\."M MISOPROSTOL STl.l)\' SATISFACTIOS SlltVE\'
Please r:ate the rollowint narements on the basis of srrontiY qne • 10 and stronfly ~isatne • I .
• . I was very sarisficd with the care we received ~urint labor and delivery. I :! 3 4 S 6 7 8 9 10
.:. Sufficient anention was paid ro the safety of :norher and baby durin1 labor and delivery. 2 3 4 S 6 7 8 9 1 0
3. The staff rave us &II the care and attention they :ould durin1 labor and delivery. 2 3 .a S 6 7 8 9 I 0
~ . Some unnecessuy inrervenrions were carried our on modter or baby durin11abor and delivery. 2 3 4 ~ 6 7 8 9 10
~- Our wishes were always respecred durintlabor and delivery. 2 3 4 S 6 7 8 9 10
6. I feel happy about this labor and delivay nperienc:e. 2 3 4 5 6 7 8 9 10
i . I fell in c:annol of whal bappened duriq labor And delivay. 2 3 .. 5 6 7 8 9 10
8. I fell some mistabs were made in the care received :Tom thc staff durinl Iabar and delivery 2 3 4 S 6 7 a 9 I 0
~ . If rhe staff had been more capable durin& labor And delivery I would have ben happier with the care received. 2 3 " S 6 7 a 9 I 0
I 0. I would be feeliftl befter aow if the Slilff had been more considerale durin& lllbar and delivay. 2 3 4 S 6 7 8 9 10
11. 'The nune pve us all 1M can and aaenrion l1nnced durina Iabar and delivay. 2 3 " ~ 6 1 8 9 10
12. 'The doctor pw all 1M anenlion needed durina labor and delivery. 2 3 • ~ 6 7 • 9 10
13. I would haw lilrad die IDif 10 haw n ; ided 10 me differently durin& llllor and cleli'"'Y. 2 3 4 S 6 7 I 9 I 0
I 4. Sufficient anention wu paid 10 comfort durin1 labor and delivery. 2 3 4 S 6 7 I 9 10
!5. I would have libd the~ of llbor and ~elivery ro have been dane diffemnly. 2 3 4 S 6 7 I 9 10
109
(
Applicant's Name: Kelly Angela Bennett
16. There was 100 much equipment used durint labor ~'1d <i.eiivery .
l 7. The surf ,.,·ere sometimes rude to me durin, labor and defi,·ery.
I 8. There were 100 many sraif or studmu involved in the labor and deli"ery.
I~- Staif treated me as if this was just one more delivery.
::0. The suif helped me ro reel like tJus was ~ very speciu event.
~ 1. The appropri~te amounr of equipment was used 10
monitor the labor and delivety.
::. There were occasions when no one e~;tlained to me what was lolftl on.
23. There were unnecessary restriction on mochen walkin& around durina labor.
2.&. The most comfonable position was used for lhe actual delivery.
:s. The lhinas done 10 the baby immediately aner binh were all necessary.
~6. I held the baby as soon as I wmted.
:7. They cried to deliver the placenra IDO quickly.
~8. I ,.,.as liml all die information needed about PfOITtSS in labor.
:9. The nurse was with me as much as I wanted.
30. I saw the doctor as often u I wanted.
31. I was satisfied wilh the way pain was relieved durin1 labor.
31. l was dissatisfied witb the way paift was relieved durin1 delivery.
33. There were too many vqinal elWirinations.
3.&. Our biM plans were ipoted.
~ ] " 5 6 7 8 9 10
2 ] . 5 6 7 8 9 10 ..
2 ] 4 5 6 7 8 9 10
2 ] " 5 6 7 8 9 10
2 3 J 5 6 i 8 9 10
2 3 " s 6 7 8 9 10
l 2 3 4 5 6 7 8 9 10
l 2 3 4 s 6 7 8 9 10
2 3 4 s 6 7 8 9 10
2 3 4 5 6 7 8 9 10
2 3 " 5 6 7 8 9 10
2 3 4 s 6 7 8 9 10
2 3 4 5 6 7 8 9 10
2 3 4 s 6 1 8 9 lO
2 3 4 s 6 1 8 9 10
I 2 3 4 S 6 7 8 9 10
2 3 4 5 6 7 8 9 10
2 3 4 s 6 7 8 9 10
2 3 4 s 6 7 8 9 10
110
( (
Applicant's Name: Kelly Anaela Bennett
)~. Rec;overy time in labor and delivery was roo rushed.
36. The nurse made tile labor and delivery a bener e~perience.
3i. I wish all docrors were as aood as ours.
38. The doctor made the labor and delivery a beuer experience.
39. I did nor ellperience diarrhea durina my induction. labor and delivery.
40. I did noc experience stomad: cramps durinJ my induction. labor and delivery.
-------------- . . -·· .. -- - ~ . .
ADDITIONAL COMMENTS:
(
~ J ~ 5 6 7 8 9 10
~ J 4 s 6 7 8 9 10
~ 3 4 s 6 7 8 9 10
2 3 4 s 6 7 8 9 10
~ J 4 s 6 7 8 9 10
2 3 4 s 6 7 8 9 10
Ill
APPENDIX E: CONSENT FORM
ft~ FACULTY OF MEDICINE
MEMORIAL UNIVERSD'Y OF NE1WF4DU1111DIJ,\NID ST. JOHN'S, NEWFOUNDLAND Alii3V6\\.
CONSENT TO PARTICIPATE IN BIO.MEDICAL
Dr. K. Bennett and Dr. D. Young
You are beina asked to putic:ipate in a researdJ study. Panic:ipation in the study is entirely volunary. You may decide not 10 participate or may wilhdraw from the study at any time without affecting your normal ueatment.
Confidentiality of information concerning participants will be maintained by the investigators. 1be investiplln will be available during the study at all times should you have any problems or questions about the study.
You underscand you have been scheduled for induction of labour. You arc aware that pn:111111mlins, altbou&h primarily used as cervical ripening agents, also stimulate labour. These mcdic:alions can be administered ci1her by moudl or vaginally.
Misoprosrol is a prostaaJandin used for the ueauncnt of stomach ulcers. Recent research has shown that misoprostol placed vqinally effectively induces labour. Our own research has demoMtrated that misoprostol by mouth is no less effective or safe than usual labour induction. This trial will attempt to dec:ide if misoprostol by mouth can be used as effectively as vaginal misoprosrol.
If you cboase to enta' dlis SlUdy you will be randomized (chosen as if by flipping a coin) inco one of two poups. All patients will receive a pill by mouth and a vaginal preparation. One of chese prepantiom will contain misoprouol, the ocher preparation will not contain any medications. Every palient will receive an active mcdicaDon either by moudl or vaginally. Neidler patients nor investiaauxs will know aroup assipment. There will be no additional cxamirwions or blood cem. After delivery your chart will be reviewed by die research team for information reprding your delivery and baby. You unclcntand that you will remain under the care of your physician who will manage your labour/delivery as deemed necessary. Your participation is voluntary.
112
You may choose to withdraw from study at any time. Prior to discharge we will be asking you to complete a brief questionnaire on the satisfaction of care during your stay in hospital and throu&hout the labour and delivery.
The alternative sllould you choose not to enter the study. would be induction via prosraalandin concainiq cream given per vagina.
You have discussed the information provided with your physician a'1d he/she has answered any questions about your care.
You undentand d1at records concerning your labour, delivery and hospital stay will be reviewed and you give your permission for this. No records bearing your name will be provided to anyone other than the research team in this study. You will not be identified in publications in any manner.
lJabUity se .. eanaa:
Your signature on this form indic:ar.es that you have undencood to your satisfaction lhc information reprding your participation in lhc research project and agree to participate as a subject. In no way does this waive your legal rights nor release the investigators, sponsors, or involved institutions from their legal and professional responsibilities.
(5.......,. or Partidpaat) (Date)
(S ...... ofWIIals) (Dale)
(Siplllun of laftldplor)
113
·Memorial University of Newfoundland
Humaa lft..-iptiaft Caamirta ~ and Craduale Studila Facuhy oi PoWidM 'llle Haith~ eaaa.
29 April 1996
""" ... Dr. Kelly&-. c/o Dr. n.ws v--. Oblllaica a O,aecolo&Y Once a.-.1 Holpilll
DarDr ........ :
1'hil will Kbo•Jedae ndipl of' your reviled COIIICIII f'orm for tbe ~ study enlided .._ .... ...,,._,, CNIIIIII, ..... •iaedTriiiCe• ...... OniVenuV ...... Mllapn!ICII PwiA ... r t.diJCIIM•.
I haw miewed tbe reviled couena form llld 6nd it to be lllil&ctory. We wiD keep a copy iD tbe IDCOflice.
cc:
St. Jolla'•· Plcuf 'Jed Clalda AlllV6 • T&: 17'1191 TJ7411• • Faa: 110111 TJ7·~
114
APPENDIX F: MEDICAL RESEARCH COUNCIL GRANT APPLICATION
........ Youq. David
~:=r=.e~OIIWI
=.tiiis- =-Youy. David Crane. lou
MJSOPROSTOL LABOtJiliNDVCTION
"'·=-mJr--7Q.tJ ..... t. Dr. Mary lflnnah
Associate Professor FICUlty of Medicine University of Toronto
2. Dr.Jermne o.a.ereau A.aociace Profeaor
17
Facul afMedicine Uaiv:.ty of Brililh Columbia
3. nr. area a,... AuiltUI Profeaor ~- ofQblf(Jya Unawnity of 'toronto
C!!!,.,, ................ _ ......... (! ___ ,
s 11,522
=:::u==:=.:-~ ......... , y_., .. , .-,.....,...... 1947 03/0t/97 1965
C·=~-~ REPR REPROOUcnONIPREGNANCY
l!·c!ar..==~
,.....,.,., ..... ..-..c., ..... , ..... , Naa •tallfiMIW
Depr. of ObsiGyn 76 Grenville Street Toronto, Ont. M5S 182 (416) 323-7317
B.C. Women's H0111ital IT62-4500 Oak St. Vancouver, B.C. V6H 3NI (604) 175-2424
Mount Sinai Hospital 775-600 University Ave. Tor011to, OnL M5G IXS (416) 516-1415
St. Joseph's Heallh Centre PO &ox sm. stn ae esc LGndGn. OnL N6A 4\'l (519) 646-6091
Cliaic:ll hMslipion
Clinic:ll Trilla
Clo!NiMe.
"''"*"'== Perinatal Ctinic:al Trials
Maternal Fetal Medicine
Maternal Fetal Medicine
Maternal Fetal Medicine
Maternal Fetal Medicine
Q.lf_.,,...__...._ ...... ,.. ... .....,_ ,_ __ MO'f ...... .,.... ........
M-e ........ r •••-- e ••- a~ 1• =• Dr. Kelly 8ennla. Resident V, Dept. aiObiGyn., Memorial University ofNewfoundJand
Dr. Rary Windrim, Pratasor, Dept. ofObiGyn., Memorial University aiNewroundland
Dr. Kim Bult.ltesident lV, Dept. of"Ob/Gyn., Memorial University of Newfoundland
1-0IN--~-·---= .. ~ .... -· ===--.............. _ ....... ..
c
c
ll5
~ c..._.~
Yogq. David
C~~DC.Io.t
.......... ,. CJAA
Maaarial Uaivenity ofNcwt-illal
Induction of Llbour with Oral Misoprostol.
=r.:.'~--..... c.,... ........... flwLiiUiii&Wt ....... :
,... ® NcM ...
lt-.1 0 "-*"·· ''~r" .,..& ............ •c01w11 .............
Mdi .......... CIICitll:
~--· ........ ==:· ...... IYN,N , ... ,,..... Y• .... Te-a. a...,. ·-..._ ~
........ ~ 0 ...-----·
__ ...... Alilllll 0 .. , •• 11 0 0 .pelllll .. l .......
A~IIW --· 0 :-c:r.-. 0 0
N":..~-....._ I =:=•-: .;; , ... ..,..d .ca~. ..... 0
a.....• ........ - .... =--- "-' ® Ia--·-- - ,...0 ,... .. • .... ----·
DEMANDE DE SU8VIN110N DE FONC11~ _,_
a.--tao.- oc
Prvf .t Clulir Obsaacl(iyn 4Cl Medicine 10 I
Aslt Pial Obsaacl(iyn « Medicine l 0 I
"'=._... ..... ~ ............ --. CJAA .. ..._ ............ ,....... .. McmoriW University ofNcwf'ouadlad
~~ lnflrwtUy- ~a:::...-= e. .. ~--
s 81,522 .... tcaCIIIIW'Nmell E.._. .... Talll s 81,522 Taq~ -=--.....- =.:.r--@10 2 0 3 0 5 y_,_...._
........ ...... .......... := ........ Dr. David Yound, Gnce General pi tal 241 LeMan:hant Road SLiobn's.NF AlE 1P9
T ....... ......... 709-771-6157 NIIIMir ......... ,_ T ........ 7CJ9.7S3-1162 EoftiCI C8lllllr [email protected] .... ......
............. ....... _... _. ... a 1ft: 4 •a• r..-® , ..... :
LJa ........... ,_.
IMC/CMI ttatvwt.t) 78iaaat= ---_,_ Office ~f Research
116
·--Youaa. David
t-~"'L-,
&:..'if =Yauy.DaYid
Crane. Joan
MISOPROSTOL LABOUR INDUcnON , . .,.J:" .. ,t:;tt .....
t. Dr. Muy Hlnnah Associaac Pral'euar FICUity afMidicine Univwsity of Taranto
~ Dr. JIRtllle DIDiaau Alloci ... Praleuor
17
FICUity ofMedicine Uniwnity ofBri1ilh Columltia
1 Dr. <ina Rya Assistant Prafasor ~ ofObii(Jya Unavwsity afToran10 Dr.RenatoNatlle
4 Pralaaar n.-otObs/Gya tJ;rvwsiay otw ...... On11rio
C!!l,.,,,... .. ~--·-....... (!!r--t s 81,522
=-~=-·CIIM ....... t y_ ...... , ............... =--' =-
1947 03/01./97 1965
t:-t=tl.\. ,.,.;_. REPR
R!PRODUC'nONIPRECiNANCY
I·~=.S:.tfa'a
e--..,., ...... ~, ..... ,....._, ___ ___ Dep1. of Obs/Gyn 76 Grenville Street Toronto, One. MSS 182 (416) 323-7317
B.C. Women's Hospital IT62-4SOO Oak SL Vancouver, B.C. V6H lNl (604) 175-2424
Mount Sinai H ·w 775-600 Univ= Ave. Tcr01rto, Ont MSG 1X5 (416) 586-8415
St. Joseph's Health Centre PO Box sm. Stn Cte esc London, Om. N6A 4V2 (519) 646-6091
Cliaical hMIIipion
CJiaic:al Trim
O.......e
=·"'*"'== Perinatal Clinical Triats
Malemal Fet1l Medicine
Ma1emat Fecal Medicine
Ma1emaJ Fetal Medicine
Mllemal Fetll Medicine
H-C ...... IIJ&._.. C: •••- • 1• a Dr. Kelly 8eanlft. Resident V, DlpL ofObiGyn., Memorial University afNewf'oundland Dr. Rary Wiadrim. Professor, Dept. ofObiGyn., Memcrial University of Newfoundland Dr. Kim 8uft.llaiden11V, Dept. ofOb/Gyn .• Memorial University oCN.wf'oundland
c
c
117
(! .. _,, ................... _~., ...... ,
Yaun& David
,..fl ............ ,,,.. ............. ___,
S ll.Sl2
Tll.NIIf .... lllrtllllllllrtNI·W.--. .....
709-71WI57
'•IT .......
709-753·1161
lt ..... <' ...... _. .......... ..._., ... w:w:~•..-1 ......... £ ............. .._ .. _.... ,.........,. ....... ,_ .... lUll.-•-==-~=-....... - .......... - ........................................................ .. ............. ifiiCMieltlla ....... ., .... .......
Labor induaion is a &equena obstdric: intervention (=201'/t). Prostaglandins (PGs) are
ettective apnts. but pstroinlestion&l (GI) intolerance has limited use to non-oraJ routes. The
tnditioMI oxytocin .. drip" requires intravenous (IV) use and discourqes mobility. Misoprostol. a
modified PG. is nwkeced for onl UtabiiCld of Gl disorders. but initiases uterine c:ontrac:tions. an
undelinble Gllide e&c:t. Recently. there hu been a research "boom• on misoprostol in prepancy
to eft'ectively and safely empcy me uterus. by usina this .. side etrea··. Almost all this research has
been per vqina. We are oae of two or three sroups worldwide who have published on oral
misoprallol to induce term llbor to IIUdy efl'ectiveness, Gl tolerance. and safety for mother/ baby.
Cost per patient has been less than one hundredth that of other PGs. even less than IV oxytocin.
This project wiU advlnce our research for term labor induction w;th two randomized
conlrDIIed trials (RCTs): one msr=iac oral to vaaiftal rnisoproaol w;th unniotic membranes ilact.
the other onl misoprosaol to IV oxy~ocin with prellhor ruptured membranes. We will assess time
to Vlliftll binb. mott.A»aby wd ' · .. motber•s satisfiction. and any Gl effects. With 6uther
...-rdl we hope to fiDd nUipiOIIUia d coli etrectiYe oral induclion qerst .
........... ,.._ .............. ..... -· ...... ----=
...... Dill/ u.• a.
S\-~.J-.1~ I ~
-----.... . .,.. ..... _, .... ,., c...•-..-.---:n_~·-====w:=a::=r .. ,.:.. ..
181 ... ,...... ....._ ... .... 181 ·::==: ....... :.diMft ....... ~
118
..... ., ..... Yauna.DDid s 81,522
.......... ...-. .... -.,_ ..... l._._....._......, • ........_ ....... ,z ....
,. ..... --• I '::Sf:tu I s 39,172 s 9,961 $49,140
=J=: I $22,600 s 5,650 s 21,250
===== s TOTAL 'IRSCINNIL I .... TOTALI POUlt Ll 'lllSCINNIL Dl RlCHEJlCHE
1 w::.-=:·m=o.:o•' I -- ' tiC.) I l a.a.. .... ,~ ........... j .,... ... l ....... ~ .. I
TOT A&. 1'WAINEES I TaTA&. STAGIAIRII
....... ,.._
.. , ~~-- ·--...... CIW.......__..,,.._If ........... l TOTAL
... .. s
s
S SIS
s II
s 2,759
TOTALS I TOTAUX
s 71,090
I
I
I I
I s
s 3,432
s 11,522
J
J
I
119
·---Yauq.David s 11,522
.._._.. ·--............. A.,. ........................... .. .............. fiiiii!L
IOhnlwk
20hnlwk
a. _,.., ..... 11 ...... .,,._ ................ ....... ___ ._ ......................... .. ..... .,_, ___ ,. ................ _ .. ,.ir= ..... .. ...... , ... _,..., •• CUft. ..... , ... ,...._.,..._ .... ................... , ..... _. ......... ..
......... (1 ....... - ....................... , ...
..... , .... l:r--'
c. ..... ., ....... &.
-··= .. Slilaylllllft
s
PERSONNEL 8-Ril 6 n'mr1
Position: a-dt Assisunt Credentials: BllcMiar ofNunina. Level n pay scale (source: Collective
AiJWW NLNU, NHNHA and Treuury Boud) Dep11 D'MIIl: Obllltrics A GyftecoloiJY, Faculty of Medicine. MUN
r.,.,.. !aR'N rs r: This is a p;ufi I" aMI polition providina racudt 1Upport to the office of Oblletrica A G7IIICDioaY· The polilion llllintainl a liaison with ofticills of the Faculty ofMedicine. Nuniftl. Pharmlcy and lfosPtaJ Adminisuation. General supervision is provided by research applicant and Discipline Chair.
.....
120
Youna, David s 81,522
Qulia: • FllftCiian u full-time project co-ordinaror. • Collect_, collllla eft0111phk:, iiiD'Ipll'tUm and ourcome dala. • lftpul dlla ., aapuw. • Ift(ann P*ftdal puticipuiS about raeardl project al prenalal
clules. • Partici.,.. ill informed consent process. • Ad~ pollplrtUm Salisfaction quatiol\11aft. • Educare nuniq and medical staff such as family physicians,
houseslllf, llld cuaoom nunes, reprdina study proux:ol. • Conduct lilaabU'e teUChes on relaled topics. • Prepare presentations ie. slides.
Salaries: Nunina (screeniJ• a: moni1orins). overtime, scandby, beeper. milcqe. IMUIIIeave S 39,872.10 Benefits (25% 1'011 salary· CPP, PSPP, UIC, WCC, etc) 9,961.00
D.ereUJ 'tr' ' ''I Position: Pbumacy Technician I Credenbals: Phlnnlcy Technician Certifiwe. Level III pay scale (source:
Collective Ajlee111ent NAPE HS, NHNHA and Treuwy Bolrd).
Deputment: ........_,, School ofPhannacy, MUN a: Health Care CCJI1MX'IIion (HCC) of St. John's
r..., eq.,. .. 'Hir: This ila psof'wllioall poliliaft pnMclina resarch support ro rbe oflice of' Obt&etrics a 0, 1 .a..,. The position m8intains alilison with ofticiafs of the Facuky ofPIIIrmlcy and Hospilll Adminisualion. Oenenl supervision is pnwided by r.-rch appliclnt, Obstetrica a: GynecoJoay Cblir, and Director of'Pbumlcy, HCC.
Qulia: • FuactioD u ftdl-dme pharmal:y project miiiiJU. • Preplle ave c1n1a 11111 pllclba u per andomizalion schedule. • Mliallia records of pllilnt usipment.
Ada phumlcy liailon with Once Hospital lite, HCC.
Saluiel: OYenime. standby, beeper, milap. annual lave Bene6ts (25% paullluy- CPP, PSPP, Uie, wee, etc)
6.2
22,600.00 5,650.00
'71.190.10
121
Youna. David s 11,522
EXPENDABLE
Total prior to taxes 50.00 HST (10.31%) 5.16 Total posa.p 55.16
Eltllultd cost of pHMifu (SW ... da). $40.00/rnouth X 12 mondls 410.00
HST (10.31%) 49.49 total phone/fix 51t.4t
SERVICES $0.02/copy x 4,000 copies 80.00
HST (10.31%) 8.25 tow photocopyins 11.2!
OTHER EXPENSES
MataialaadSuppiJa
Trani
Estmlated cost et •lleproecal (Rcr 1). SO. 77/patient x 206 patients I 51.62
HST (10.31%) 16.35 tow misoproaol (RCT I) 114.9'7
~COlt etplaah (RCI' 1). Sl.OO/paliem" 206 palients 206.00
HST(l0.3We) 21.24 total pllcebo (RCT 1) 12'7.24
Eltbuted COlt ............ al (RCI' D). $0.77/paaient X 54 patien&s 41.51
HST (10.31%) 4.29 tow miloprostol (llCT D) .t5.17
Eadluted c..a.r • .,... (Rcr m. SS.l2/pllient X 54 palientl lll.U
HST (10.31%) 29.06 toal oxycoc:ift (llCT D) liO.t4
EftieatM celt rortnml r.r p.-tatiH et resulas aad uplou1 r. .. n '*led ,....a epport.aitia. z.•.oo
TOTAL EXPENSES saa.sp.n
6.3
122
......... --..___. ...... _.,..,,..._ .. _...., ·----Ct--. s 11,522 YOUI!IL David ...................
. !..... =-:r.._.. ...................... ........ - .... ..,._ s......, ofDflcuuiH
c .... '-----.............. ...... u...... .. .. .....
1be conniaee wa c:oftCIImlal that misoprosaol wu not beina compared to current ltUidard of care. c.vicll prullllllndin (PG) widl or without IV oxytocin. and that thouah labor duration wu an outcome of interest. safety should be the preenUent outcome.
AI is stated in our research proposal. we..,_ thalsubstantiw nwemal (e.J. Caeslrans) and newtan (e.a. birth uphyxia) are the prd'ernd primary outc:omc meuures. however, the lllftllle siza ...ted to 1pp1upialdy evaluale clinically impoRint dlanaes in these (see RATIONALE PIP 12.6) would require 3400 and 5030 subjects respectMiy. Such a c:linical trial would need to be multicenter and. lhere(ore. cosdy. Belore bep.ail• a naalticemer RCT, the most appropriale misoproscol interveftrion needs to be defined. Oral misoprostol would offer a new approach that would appeal to many pelients and Clfe8ivers. The RCTs we propose here are to evaluate an oral nd:lpiDihll ptatocol in labor induction qainsl V11inaJ misoprostol with membnnes intact (RCT I) and .... .....,.oxytocin with prelabor-IUpCW'ed membranes (RCT U). We feel our public:alions lUI
haw ahown Vllinal rniJOIWOilOI u c:osa-eS"ective c:ompared with a cervical PO/IV oxytocin protocol and ani misoproltol no less. lfthe proposed orallllisopro•ol RCTs suppon existing evidence. then a multic:emer RCT of sufficient sample size would be tppropriate. with a primary outcome such u Caesarean or birth asphyxia. and versus the cumnt standard of care.
Rnllwer II "DIIed Dea.ber 9, 1996"
Our proposal has been extensively revised (RCT I and RCT II). and we believe is bener orpnized and not repclitively presented. The demonstration of a cost-effective and safe onl induction apnt is clinically relevant. The other two reviewers support this. Peer miewers for our oral misoprostol publk:ldon 21
• have sugesaed that this may lead a dwlge in saandanl of care in NOftll America. after fUrther evidence.
Onl milopioal is likely to be leu c:osdy even than IV oxytocin induction. where our co•s ror IV liae. ftlbint. e~ecuo~y~e solution and oxytocin awnae ss.22 per pllient.
'Jbec..diln IIIINI&charerofnUoproslol (Searle Canida. Oakville. Onlario) hu not been interelled in IUppoft of research in labor induction or prepancy tcnninalion. This relates to IFW& I 1 ' llbilicia inbennl to raan:h in pi....,ICf, and muketina liabilities ofbeina IR«'Ci••ed with a potllllill'"lborlion" dNa. Even if misoproltol became the ideal pharmlcoloaical induction ..- and the fine line appfOICh, the multet size and lddilional pnerated revenue would be small, coaapaed to the present ..r. and secure Gt market. Unless the company adopcs a .. for the aood of women" laamaniwian attiblde. it will never be proiCtive in this uea of research and IMMr .t Hllllh Procection Branch IPprGval for investiplion of this new indication. Our research hu been wowd by CU" umncy llld ....-. human investipaion llld edlics committees. and subjecU ue filly nbrnwl W'dh Airther ~ e\idalce we would leek such HPB IPJI'Oval, in defauace 10 die IIIUU&ctunr. ['TheR is a rwnour that the company may beain mulcetina misaprolcol oaly in comlrinarian with a nonsteroidll Ulli-in8ammalory aaent in a siftale piU (AilTHROTEC). which would eliminate use in prepancy.)
..
1.,~ _.,
Youns.David s 81.522
A two-tailed sample size calculation was used because eft"ecc may be in either direction. Excessive uterine activity vuiables wiD be coUected. u well u substantive newborn outcomes. Oxytocin use in our study and control JrOUPS have been specified.
Ia lbe put yar we have been invited to lftl8ll our raan:h at national (SOOC) and reaional _..in Cllllda (Adallic and Ontlrio).llld It IMnlteniuy c:eaten • Calpry, Halifax, Toronto and Bllrimore. Raean:ll with misoprostol in obstecrics Uld IYftMCOIOIY in Canada is underway in V111couver llld Toronto. I hope this woufd persuade this reviewer of the general interest in the Canadian obstetrical community reprding misoprostol.
Reviewer 1¥2 "Review of Protocol by YOIIDJ D. 9609 cr -34561 • cr B"
This review was quite encounsinl to us. llld made a number ofbeneficialsugestions. The reviewer felt that oral misoprostol is "a potentially imponant addition". and that "misoprostol may became the tratnlellt of dloice" for induction oflabor. Our difficulties with obtaining funding from the manu&cturer and through MRCJPMAC were understood.
Concern reptdinJbreWy ofthe resareb plln. clarity of inclusion criteria. and stratific:arion wriables have aD been addressed. We have not specified a maximum Bishop score. however. u we are explorina misoprostol use for the tWI specuum of cervical .. ripeness".
We have darifled how side affects will be assessed. by c:aJe11ivers and subjects, and referenced our Jatil&ction queslionnaire.
Because, foUowina revision. the projecc will be completed within one year, we have not atllllished an edemll Safety Monitorinl Cc1rnrNaee, however, any in11ance of neonatal asphyxia will be brauafll to llllnlion of die study team within 24 boun of its occurrence. for ongoin1 rnonitoriJI&.
'I"M reviewer hu • ...,....., addition of a trials design expert to our team. Dr. Joan Crane is now added a a ~principii imoeslipaor. She is alllltanll fecal medicine subspeciality graduate wflo hu now joined our ficulty lnd is at present~ her M.Sc thesis in c1inicll epidemioiOSY at Dllhaulil. Tile two colllboratil• .aior resideab, Da. Kimberly Butt and Kelly Bennett are enroUed in the cliaicll ~ M.Sc prup~m at Memorill Univenity ofNewfoundland. The resources cltbe C&aical Raeardl Group in C1inical EpidemioloaY at MUN are available to us. The principal invesciptor hu 111 M.Sc in Desian. MeuuRment, llld Evaluation fiom McMaster University. Dlputment of Clinical EpidemioloiY and Bioswislics.
Reviewtr 13 Oa a ro,. eadtled C.._ittee Review's Report
We were pleased wilh this very positive review. Throup the revision for resubmission. our .... a. bu t.n reduced so that completioft widin 1 yar u a sinal• c:enaer study is now !euillle. We f'eel that followinl this project we will proceed with a mulricenler trial u mentioned above.
I Wll deeply traubled by the &nil line clthis review. It is.....- that the reviewer's oripw raq Wll •excellent'" with a score of' 4.4, but this is penciOed over to '"aood" and 1.0, with no iDdicldon u to who made this chanp. Presumably the committee hu this autflority.
9. l
124
YOUIII, David
._,.,._. ....... =~~--==----· ..........
...... ..... ::-.:-':-J.:.:=.====-··-
Llbor induction is a &equent obstetric ilarlentioa (•20%). Prostqllndins (PGa) .. e&1:1ive......, but plllodalilill (GJ) inlolaanc:e hu limited their use co intr.cervical and vqjnal ldninillnlian ofPGEa aels.
Miuprual. POE. lftlloaue, is rnukelecl Cor oral treabnent of upper Gl disorders. The put tMt yan hu - mUnomifta Utmaure on iu .-to initiate uterine coatractioas f'or prepiiiCY tan· •;,., n in the &It lnd .and trimestas. and llbor induction in the third. Vqiaal administration hu bem llmolt m:ln6ely Ulal. hll beal ccsc4'eclive (lell than oae hundredth PGEJ and without hlnn co mocher or newbcm. We have published a randomized comrolled erial (RCT) on vqinal use ll. An onlllbor inductioa 11ent would be ltti'ICiiw to many patients and care providets. A sinaJe ptE hldlt.CT IIIIIIUICripc 11hu evaluated onl 11ilap011d u • one dose cervical Primina aaent prior to oxytocin induction, and our RCT" (in press) with 27S subjcc:u addressed an induction prot~cJI with repeat oral doses compued to a nditional induction rqime (physician chosen combirwions of intracervical or vaainaJ P<i~, intravenous (IV) o"Ytocin. and anific:ial membrane rupture). Both acr·s found oral misoprottol effective. well toleraacd. and without harm to mother or newborn.
Bef'ore a.._kiic on a costly and necessarily ndticenter R.CT to evaluate more subsuntive outcomes (Caauans or neonual asphyxia) with sample sizes pacer than 3000, we seek tUndiq fbr 1WO RCT"s. fi:lr labor inlb:lion 11 term. RCT 1- • double blind comparison of oral venus vqinal niqlroltul, widt intact UGIIi)nws and R.CT U - oral misoproSiol versus IV O"Ytocin with prelabor ruptured membnnes (PROM).
~ ltaaldt Quldoal RCT I · When COIIIpwed with vqinal rnisoprcmol (SO~~ ~ 4 houn u needed) for labor
inlb:lion • term with illlaCt ••••a•w:s. does ani adminislntion dift'er by more thwJ 4 houn in time &om induction to vaainaJ binh?
RCT D - When compared with our escablilhed IV oxytocin llbor induction prot~l for term PROM. does oral misoproslol (dose a above) difFer by more than 4 hours in time &om induclion to vqinsl birth?
Secol•y researdl questions address harm to the newborn (mcludins cord blood acid base ...,. a ACOG birth uphyxia crileriaa.u) and mo&ber (Caesareans. peripartum interventions). mMemll Gl inlolennce. and excessive uterine ICiivity.
Raardlftu Subjca in bolb lt.CT"s wil be 11 psaltionl.-er dian 37 completed weeks. with a cephalic:
presenti.., 1M linale &:Ius. who bave an iacfic:laion f'or induction. and no c:anuaindic:alion to iaduc:liaa. Yllinal binh. or PG ua
Rladam elloee«m wil be bloc:Jced and Slnlificd (on Bishop ICOI'e Uld parity). In RCT I, Ill IUbjlcll wil naiw a vqinll Uld oral applicalion (misoprollol or p&.cebo). In R.CT D. the conaal lfOUP will rwceiYI IV oxytocia, bued on recen1 TER.MPROM ma&JU'. Blindins otR.CT 0 is not ttuil* for our center • present.
Semple U. C'h""DMD were baed on a • 240·miruta. ca (2 tailed) • o.os. p • 0.20. wiah o &am our pt& "o•lft2L· Adjustment Cor lnlic:ipad c..rans (<20%) were made. Sample size for R.CT I is 206 and RCT D is 101. This recnaitment within a year is supported by our prior resarch.'w
·•
125
. ------------...-(111,_,, ...... _ ...... y GUlli. David
•-...... CP-S 11,522 (......, ........ _____ ......................... _
,._..._ .. ..__,..._.IIIIC ... f.&-- .._., ....................... :-.-
::.=:- ·-------· ==:.-:::.-r.. -·---____ .., .. _ .................. -. Our pnviOUJ wort or 111011 direct relevance to this application hu reaalted in tbrw
pulllicllions in the puc ,.Vt.1'
21 • These papers ,..,an our two prior randomized c:omroDed trilla
(RC'I's) al\'llinll1• IIIII c:nl milup0110121 wru a widely esubJished tenn labor inducdon pratOCDI. IIIII a liclralure ~'~View" IIIII ........ayu ofvqinal miloproltollt.CI's. These papers are llucribed in the lt.aan:h Proposal. paps 12 to 12.10 ofrhis applic:ldon. with the complete papers in the Appendix. Data on vqjM1 binh interval (VBI) in the membnna intact llld membranes rupcured (PROM) scraaa of our onl misaprostol publication:• will be preseaaed and discussed here. Some or this clara cloa not app11r in the paper, but will be praented at the SOGC ACM 1997.
Of' the 275 IUbjeetJ I'IJiduuWed. 56 were in the PROM llnlUm. VBI (cesuans aducled ftam lllllysis) was an-. (a SCindard deviuion) of7l4 :t 461 min with oral misoprosco1 venus 557 :t 312 min for control subjects (1V oxyrocin) (P-0.13). Nonparameuic analysis wins the Mann Whitaey U test fancludina c:aare&lll u the same VBI. but pater than any actual VBI) Jives a median VBI utiOO miD with onl misoproscol and 796 min for cunuols (P-0.96). No newbom in either poup hid binh uphyxia., •.
11oe intiCt nwuDw•lb'IIUm coaaiaed 219 subjecls. The control poup permitted phylician c:tm.n combinllions af~ prosllllandins.lrtificial membrane rupture. and oxytocin infillion. n. VBI (c:e.•as excluded) wu a mean of974 2: 524 with oral misaprosaol venus 1002 a 606 min in controls (P-0. 73). Nonparamctric analysis fandudins cesuans) siva a median VBI afl030 min with oral miloplostol and 994 min with controls(P-0.34). Apin. no newborn in either IJ'OUP hid birth aphyxia. 1ft neither sanrum wu a dift'erence in VBI found. Even direction in eft'ea time is not tpparalt.
Our vaainal milaproaal RCT .. for tenn Iabar induction included only intact membrane .... =v &y cm.ia ... euentially the same u for the intact membranes straCUIIII or our onl milopiCIIIIDIIt.CT21• W'11b doe control poup (as for intact membranes scraaum above) mean VBl wu 941 a 506 min (excbina ccsarana). with a median lime of931 min (includins cesareans). Very 1imiJar clara to 1hll in till prwvioul puaaraph. Vqinal misaprostollead to a mean VBI of'753 :t Sll min. wilh a median lime oL611 min. Campara~ with the cunuol poup, these VB Is durations are less (P-0.011 t-tat, and 0.017 U-tat rapectively).
'l'lil.-dl.,..-will describe two RCTI. RCT I will be a double blind compuiloa or ani venus vqinal rMDproaol for term induction with inciCt neubriDD. It would be loaillic:IQy noon ciUIIcult. if not q a 1libll far our cent• to carry out a double blind RCT or oral or wain~~ llliloprollol Yaaas Uldidonal induction melhocls. without placebos provided by phamaceutical IIIUUfactunn (Sarli Canida far niluprolao1 and Upjoluo for diaoprostone). which is not fonlscomina II ,..__ The above dale .,.aas onl misoproslal .-.Its ue very ....,. to die c:oattol ..., ,..liadarly with nmbranes intact. A double blind trial of oral venus Vllfnal miJoproaol is f'eaibll for us and would pn:Mde unbiased outcome data. particularly on the mon IUbjectiw uterine aaivicy.
RCT U will be clesiped with sufficient power to come to a conclusion reprdina oral llliiapiOIIOI wrsas IV oxytocin (no better method yet~- Blindins of this RCT is not feasible forUI.
\
.. ,.
126
,... .. ..,. ....... ....._ .. _ ....... .,.,..,, ................. YCJUIII, David .. • ...,1T:!!(1"' ......
------·--·-::r.·-·--·----· =.:.t;,;t,Jf ..... ----·---~ &.....-.. - ....... 0 ._. ........
Title
Induction of Labour witb Oral Misaprostol.
lalndiiCtio• of Labor wi .. l'nltqludiu (PGa)
........... ==--=-~-.............. _. __ -----~===== ----~ . ... . ............. '-__ _ ---------
Induction of labor nar term is fiequently necessary. (more than 20% of labors in many Canadian tertiary centen) for 1 variety of obstclric and mediQJ indic:alions. PGs. panicularly dinoprostone (P(i~). have bela prown effective c:erYical ripenina and labor induction lplll. u Since 1971 there hu been evidence from randomized c:onuoUed trials (JlCTs) that oral PG &dministntion is an efFective merhod of labor induction. 2 Unfortunately. oral POs. includifts dilloprostone tablets. lad to uucceptallle pstrointestinal (GI) side efFects. with up to a 101.1. incidence of vomitifts and diurhca2• Wilh the development of commercial tel PG preparations dllial*l for Vllinal or intr~~C~enical application. without such side etfects,the oral PG approiCh for labor induction has been virtually abandoned.
Milo .........
Miloprostol (~ Sarte Canada. Oakville. Ontario. Canada) is an inexpensive synthetic PGE, analope marketed in North America in an oral tablet form. which is liable It room tempenaure. Two formulaaions. I 00 tal llld 200 talo are aYiilable on physician prescription for pm•IDon and bllln•ll oe ..... II'Oidll lllli-irdllmmaa dNJ induced pstric ulcers and b'atmem of .. ,' .. u1cen. 1 ne hqueacy ofGI side efl'ects is low with oralldministnlion of up to 1600 "' per clay. U• in PI¥W"Y is Silled to be cofttrlindicated because of its uterotonic etrects and the lilt otauiaaai;p. Then bu been no evidence offctotoxic. teruopnic. or carcinopnic etfecu in animal studies. 1 Misoprollol COlli less than one hundredth the price per dose of commercial PGs curnndy used in llbor iDduclion.
A role for Vl8inll miloproslol has t.n established in second trimester terminalion of pr..-r:). U1ina its utaocoaic e8'ects u in divided doles oC 400 to 2200 ,., per clay. An RCT hu canpll'ld llliloprostol &vanbly to clinoproaaone•. Our JRIUP &rst studied misoprostol in second trimester tenninalioa ofpi.WCj c:omplicaled by lethal r.taJIIIOIUiies, and found vqinaiUie I
co•-e&c:Dve to intrumniocic llypertoftic aalifte.' Misoprostol is allo eft"ective in ftnt triiMMcr llldcll ws;Waau. but only folo ... ailistration of another abortifacient. such u mileprislone' or methotrexate'.
v..-.a ~.a fer La1Mrl8d•ctio•
Sewral RCTs have now shown vqinal plKemcnt of misoprostollablets to be an cft'ec:IM lllllhod ofua.cina llbor. wilhout Ul increue in adverse maternal or neonaw outcomes ... ,, In 1993. Slnchc&-ltlmossal. reponed • RCT of ,..maa misoprostol venus intravenous oxytocin ildbsion with dea wd time to v.p.l cleliwry. Subtequent trials have compared a variety of controlpoups
127
... s 11,.522
iri ... ilmcervical and vqjnal dinoprosc~••. In 1ft RCT iAvolvina 222 subjects at term with membrlna intact. our JI'OUP'• fouad deaaled time to Yl8iNl delivery (by approximalely lhne baun). lela hqucat ~ ..... Ilion, & ~Croft~ trald (or less use Of epidunlt'lllpPI, end 110 difl'awcc in Caesarean births witb llliloproslol. Median PG cost per patient with miloprostol wu one lluadnldth lhlt in the COIIIrOI tubjecu, who received our established induction protocol (physic:iiD-chosen combinltiou of iall'lcervical or vqinal dinoprostone every six houn, anilcill rupture olmembnna (AROM). IIIII oxytocin infiasion).
Our poup reponed ellldl-lnllysis of the published RCTs (S papers"u. l msncu•••• and one leur') or vqinal milaprollol venus c:ontemporuy comrol sroups (most indudins vqiall Uldlor cervicll commercill prostqlendifts). Over 600 subjec:ts receiving vaainal misaprostol hid been observed prospectively in such research.11 Misoprostol hu been effective. Tbcre wu more fiequat vqiall birth within 12 hours (Relative risk {b) 1.60, 9S% confidence intei'Yil (Cl) 1.21 to 2.06) end within 24 hours (ltP. 1.16. Cl t .ll to l . .Sl). Oxytocin use was less (lUl O . .Sl, Cl 0.47 to 0.61). No sipificant chlnp in Caesareans, low five minute Appr scores, or rate of neonatal acidosis wu (ound.11
Oral Milopr01tol ror Ll.._ ladecdea
An orelly administered Iabar induction qent would likely be annaive to both patients and laJib cere proviclen. Awidlnce ofimwnouslines in some panurients, and less frequent need Cor wain~~ aa1inll:iun miaht be llllic:ipated, and considered more client fiiendly. Greater fieedom Cor upriaht positionifta and ll1lbulatioD miaht even ficiliwe labor prosress. •
Beceuse misoprosrol is ~ to be well tolerated orally when used for its primary ildic:llioa. the nw 1ement of upper~ dysfimc:tion, we studied the efl"ecriveneu, alety ...Slide dcls ofmiloproltol• • on1..- for inducdon oflar. Althouah no infonnalion aists on cnllftilapiOIIOI phll1lw +" "'" • in third trimester prepu~CY, first trimester uterine contrlelion ... fbllawina cnlldmiiUIJIIioa suaesc responsa simiJer to that in nonprearllllt use. After oral adminisllatioft in mila, the peek coacentnlion and half time of misoprostol acid. the ICIM meubolite. are 12 end 21 minutes respeclivety».
We hew in pcesa2' 1ft ltCT ofonl misoprostol (SO ~• every four hours if needed) venus our lllndlrd eppte&h to tam libor iniM:cion {Jihyliciln chosen combirlllions or intrKerW:&I, or vqinal dinopiGIIUIW. clUe iiba..._ aa,racia idbsion, end AllOM). a The 275 women were I'1Ddomizal inlo ... bued on IMIUINa•--. Prellbor Npture of membranes (PilOM) bid oc:cumd in 56. In .........,, the time &om iadiiCiion to vqinll binh with oral misoprostol wu not sipificlndy dili!r.-lam dill wilh our ab"'""'d pniCGCOis. There were no clinically or Slltistically sipiftclnt di&iaw in lllllemalleCIODdlry ~ meuura (c..ran ...... &equency or epidural ..... paialll 11'1111111, or ....a naiiMI oftbe plecenta). Neonatal outcomes, inr.ludifta cord blood ICid .,_.....,,... na& ....._ 'nere wa no di1Ferwcc iA fiequalcy ofnwemal Gl side eft'ecls in the two 1f0UP1. Onl miloprallol may be a are. cost- effective iltemative to standud induction ..-.with a hiah c1esree or,... •= c ptab:dity.
We 1te unaware of adler publicllions usins oral misoprostol for term lebot induction with livinl fetuses. Npi et II u have ncendy reponed a double blind RCT with a sinJ1e 200 ~• oral
11.1
128
•r---------------------------------------------------------------Youna,DaWI s 81,522
miloproiiOI dose vtra11 piiiCibo for ccnic:al priJnina. in PROM • tam. Twelve houn W.. m iallaavenoua oxytocin i-+"C'a protocol wu bepn. i!tbe putic:ipul was noa in proaaeaiw labor'. ,.,.._IUIIjtcts receiwd onl miiDprostol. and 41 placebo. The Bishop score wu sipificandy in41o,.. widl nilopla.al (M).05). Tliny-fbur women Biven misoprollol went into labor without oaytocin. woupaed to 20 af*-._. piiCibo (P<O.OOI). lnleMI to anset of uterine ICtivicy llld diiMry wa bodllhorter wia Dloproscol (P<O.O I). Therw were three Caeslranl in acb paup. 'Neanllll CIUialmll and Gl tallnnce _.. comparable. The lllthon concluded that a sirlale 200 ~• mitoprollol oral do• wa d'ediw for cerW:al primina. and may be eft'ective for labor induction.
The onl riiDpiOitol piUIOCOIIIIII its purpose in our IIUdy .,. dift"erent &om that or the Rona Kaaa im ·a• s. 0.. ...._ CUIIIIIIIIive dole per IUbjec& for labor illimilar to dlcir 1inaJc doa It is reaauriftl that they did noclnd alipi&cant problem wfth acasiYe utcriae activity.
Ius lbllnct hlljUil ...... lom V~ ora double blind RCT with oral misoprostol (50tal fNf1Y 6 houn) versas v.p.l clinoptostone (2mg every 6 houn), with study medication pven up to 4 doles. They found no cHaau in induction to delivay interval (37.9 z 41 hours venus 36.0 z 75 houn), bul failed induclion was biper in the misoprostol aroup (16% versus l-Je. P<O.OOI). These induction to delivery in&ervals (not stated if cesareans were included in this analysis) are obviously lonaer than thole -.. 11 our center, so we eqerly await their complete manuscript pubticalion.
The median PG COli ia our previous saudies1w wu SO.Jl for misoprostol and a conservative S70.00 for saudard therapy. Maximum PG exJ1CftSa for any sinale pllient were under S 1.00 and more tblll S110.00 for ......,.ol and standard therapy, respectively.
Fetal Saldy luaa
..._.. in labor induclion il die potential for excessive uterine activity which may interfere wilh utero: I ..a perfblioa lldl ci •ly to compromise the fetus. Becaule of confiasion llemmina tam vRid definitions of ·w Uleriae activity, CUftisD recommended a Slllldlnl tenninoloJY: h):patuuic ·a COIIIracliaa llllill mare.._ 90 seconds; tachysyllole • contnction hquenc:y anater than fift per ten mUuta; ll)pa • · •haiM • hypertonic contriCtioa or tachysystole with llle &tal halt nae clecelerltians or filii tldlyCIIdia. While t11i1 ta rnilioluai is m obvious imprcwement. it is 111\er* r'n lltlilrlly, willa • .....a to support it based on fetal or newborn outcomes. Thouah tbl; ' · lllip .,..._. • "'• .-..w&ic fetal han raae (FHR) IIIONtorina and fetal well beina is DOt one to one. ~ chan&a wociaaed with excessive uterine activity wamnt i•lltl1lion ect.to-.. IIIIRie ..mty, provide more direct uw 11111t of fetal well beiDa (f'elal blood amplina, adler biaph)'lic:al •aasnent), or removins tbe fetus liom this environment (dllivwy), if simple ......,.such • position change or maaemal O"fgeR .. pplemencllion ue not QNieeli ....
Our metHIIIIylis" of dill hm cxiltina RCTs hu found tadlysystole sipific:anlly more hquent with miloprollol(la2.11. Cll.53 to l.ll), wfth uterine hypastina&laaion MUty 10 (Ril 1.84, Cl 0.98 to 3.47). TNs lllltHnalysis included every vap.l misolJrostol protocol The IDIXilllum in it a ol tadlysyiiOic in any sinal• report was 37 pen:ent12
, and uaeriae h)peuf ••Ma 11 pen:aa. To place this in perspective. conaemporuy lhldies with oxytOcin induc:dons report tldlysyslole and uterine hypentimulllion of similar or pater ftequency.M.2t A
12.3
129
... YOUftlo David s 81,522
._. RCT report» &am die 111111ILIIhDnq fbund las llehysystole {7%) and hypentimulalion, than with a dinaprostone vqinal inlat. on dlanainl &om their oripw12 every three hours to ua every f'our hours proeoc:ol.
Tbe Vancouver double bbd RCr• found uta'ine hypen1inulation rates of 5% with oral nilupa...U and 10% with vqiJIII clinoprostone. with f'ccal distress in t.bor oflJ% in both poups, ad c:euran rues or25% llld 22% (NS) -all reusurin& results for oral misoproscolllfety.
When oxytocin inducdon results in uterine hypentimulation with nonreusurins FHR chanaes. dilconainuina the inftasian can reduce oxytocin blood levels quickly. Use of vqinal dinopronone pis or misoprostol is more problematic. Authorr' describe vqinl1 lavqe llld nmovina tablet remnams. however, intravenous p - adrenerJic qonist reaimens have been usecfl wilh....,... bene6cilllocalytic .....-. flllraw:naus tocolysis ia die apprcsch necessary with oral misoprostol. fon:ed emesis would likely be indl'ec:tive, and distasteful. We have never used intravenous tocolysis in more than 400 misoprosaol inductions.
Despite these c:oncems ofpoaible uterine hyperstimulation. subaantive worrisome newborn outcomes (neonalaliCidosis, rneconMn aspia-.bon. and ACOG Binh Asphyxia criteriaz-.n) have been l"'l'e, • tJ IIUe. and nat clif&renl bec..n nisoprostolllld control groups. It is imperative that such ourcome cilia be c::ollecled in tblure tiUdies to quancitate more preciseJy potential risks of misoprostol (and trdtionalil)proKha).
Pnlabor R•ptan oiMeallnMI (PROM)
Approximately 1% of prepncies present with renn PROM. A Ianser PROM duration is thou&hl to be ala ci1rell with a tiat-risk oflnllemll and fecal infection. Over W/e of these women will beain labor spontaneoully within %4 hours. llld more than 9S8fe within 72 hours. The etiolol)' olPROM is not well understood. Ja
Tbe usasment of suspedld PROM involves history and physical examination. lab testifta, lad ultrllound iJn1P1. A •erite !piCUium eum demoiiiUI!ina ftuid in th: vqina. which is both nitnzine posiliYe and f'crnina poli1iw. is diapotaic. N""mazine will detect alkaline lmlliotic ftuicl. whlnu the vqinal pH in PfiiiiiiiCY is 4.5-6.0. NitriZine can be falsely positive in the preiCftCe of blood. temaa. vqinilis IIIII .,,.. ; IE c IGiutions. The typical fcmina apparanc:e of ~~Miotic 8uicl, wh.• a fluid drop dries Oil a llide IIIII is viiUIIizld by microscope, is less often fibcly positive. 1J11ruound can quulilll8 innulerinl amnioeic Ouicl volume. If' doubt remains followina these imleltiplionl. ab• •an fiX" llrdlcr luid leabae is appropriale.Ja
Cancnl wilh IWub ... n.OM .-e related to ua incrased risk of maternal infection (IUCh cbaf • · iii~ pall-plltUID eradamltrilia lnd sepsis) lnd more seriously, ncoftllll inf"ection. Early induction to avoid tt.e CIDflllllic:.aions raises worry about increasin1 Caesarean births. There is cwauwny lbaul whllber to-...:. i111 5 ely or m1ft8P expectandy; or whether to use oxytocin or PG, ifiaduaion is cbolen.
Sevtnllhldielliave ___.induction with oxytocin or PG's vs expec:unc rnanqerncnt. Thae IIUdies have been small and ,.......ly biued by the method of randomization. Neonatal outcomes were not IISated blind!y. Overview of these trials wu carried out in the Codnne Colllboruion Dllab .. • Prepancy lftd Childbinh Module. Induction aflabor with oxytocin wu
12.4
130
s 11,522
IIIOCialed witb biahcr c..... rata (typical OR. (95% Cl): us ( 1.37-2.50)) and toww rilt of MOIIIIIal in&clion (typical Oil (95% Cl); 0.27 (0. i2-0.59)) than expectant muapment. D In aia11 c:omp~~iaa induction with PGs with apecllllllllllllpiiiCftt. the 6ndinp .. ..-ed PGs would f8lll in a similar C •• l'lle (typical Oil (9S% Cl): 0. 91 (0.65-1.50)). and lower l'lle orendanwbitia(typic:al Oll(95% Cl): 0.31 (0.20-0.70))~t OveMews or fROM trials complrinJ oxytocin VI vqinll PGs IUIPil that iaducliaa of labor with PGs results in fewer Caauelnl (typical OR (95% Cl): 0.62 (0.42-0.91)».
acc.Jy the TEJtMPROM TriaJIU' was published. This remarkable slUdy included 5042 pllilnla &om 72 centers in Canada, Britaa. Australia. Sweden, Israel. and Denmark. Four poupt were compared 1) induc:tioft with oxytocin. :Z) induction with vaainaJ PO~ pl. 3) expec:tant care ~ induction with oxytocin if needed. and •> expectllll care ~ induction with POE2 ael if needed. The primary outcome wu IIIICialll1 infection. AD in&nts were to have a CBC and blood c:u1ture performed and clinical "'" mcnt performed blind to lhe ueatment allocation. There wu no clitference in the rates or MOftllal infection (:Z.O - 3.0 %) or Caesarean secti\ln (approxima&ely 10%). Sample size had been chosen to detect a 50'.4 reduction in an anticipated baseline neonatal infection l'lle or 4%. Because the Ktual baseline rate was less than 3%, lhe study power was un!ortunasely only 25%. The time to deiMry wu shortest in the induction with oxytocin sroup (P< 0.0001). There WU I tread toward redudion in chorioamnionitis and postpartum f'cwr in the oxytocin induced paup. OIMr secoadlly ouacome analyses .. gated less neonatalllltibiotic use and leu hquent neonatal inlenaiw care ldmiaion ror more than 24 hours 'with oxytocin induction. Thouah the numbln inwlved are small. aD f'our non-anomalous \)erinatal deaths were in the apec11nt poups (P • 0.125). Pllilnt lltiJ&ction wu hipest in the induceclpoup~. In one public:alion. the authors co heeled that iaduction with IV oxytocin wu the preferred choice. :n
Anpably the plliealaf•fertion llld ~Kandary outcome analysis orTERMPROM sugac the approach ofi:JI 1 r.,. induction orlabor is unsurpassed. The option of oral apnt for llbor induction in tbis sihlllion bep fUrther evaluation. The existing published infonnalion fi'om Haft~ Kona 22 and our PROM stratum n is very preliminary.
RESEARCH PROTOCOL
........., lteHarU Q•~
This proposal seeks lladina ror two randomized controUed trials (R.CTs) to answer the f'oiJowina:
RCTL
RCTD.
When compand with vllinal misoproSiol (50~ every rour. hours u needed) ror iDduc:tion otllbor at term with irlaact mcmbrues. does oral idminisaration differ by more than rour hours in time liom induction to vaainal birth' When c:orftl*'ed with our eaeblisbed inlravenous oxytocin labor inductian prococol far.....,.. rupcun of' membranes at term. does oral miJoprostol (50~ every fOur haun u needed) dift'cr by more than raur hours in time &om induCiion to vqinal birth?
12.5
131
YOWl& David s 11,522
I. Does onl misoprotaol induclion ofllbor increue hann to the newborn (u mcuured by cord blood acid bue lllllysis, and ACOG criteria ror birth uphyxia) or mother (Ciesarean. peripanum UIUIIIa or iataventioas)?
2. Does oral misoproscol increase maternal Gl problems (vomiting or diarrhea)?
3. What is the hquency of uterine tachysystole and hyperstimulation usoc:ialed with the scudy induction protocols (u well u in a poup of term parturients in spontiMOUS labor at term)?
Ratioaale
Our prior research has demonstrated cost-effectiveness and suuested safety first of Vl8iftll rnisoprostol and dlen oral misoprostol in comparison with an established and widely f'oDowed labor induction protocol. Before considering a more costly and ambitious multicenter trill on a more subslantive outcome such u Caesarean rate (2N • 3400 to detect a chanp &om 12 to 9 %) or neonalal asphyxia (2N • SOJO to detect a change &om 2 to 1%), we feel fUrther raearch (&n euence pilot studies) is required to beUer define the appropriate interwntion to evaluale in such an cxerc:ise. Oral misoptosaol induction would be most innovative.
The first RCT here pro.,ased wiU provide an assessment or oral misoprostol apinst vqinal miloptollol u the initial approach to term labor induction with membranes intact. There il no •ch publilhed trial to dale. A double blind comparison will penni& an unbiased usenmcnt of aU outcama. includina the occumnce or ex 11fve uterine activity.
We feel the SKOnd RCT, with a clift'erent control poup is warranted. W&th Npaand lllllllbrana, vqinal piiCirnent, Mention, llld llbiOipCion ofa misoprostol tablet could Ill be pnlblemetic:. Our in1erpnlllion ofthe TERMPROM trial prompts us to choose traditional iaaav•IOUI oxytocin induaion for the comral. A double blind approach for this trial is possible, but not fasible, at this time in our center. ·
RCI' I· Onl verna Vqiul Miloprostol (Do•ble BUDd)
S..ple Specillatioa
Subjects wiD be ncruited &om those patieall who are placed on our hospital induction lisa &lr 1ft obsMtric or medical indication EJialllility criteria are preanam women upiUiicHl paur .... ]7 cmmpleted weeks with ........ intacc. and • sinal• live fetus or cephalic praentalion.
Exdusian criteria are nonrassurifta feW ban rate (FHil) tndr.s. prior uterine surpry. lldla1 or lllljar fetaiiiiOmlly, SUipected c:llorioamnionitis, known hypersensitivity to misoprosaol or otber PGs. or c:ontrlindic:al to vqinal birth.
l2.6
1 ... ., _,_
~.
---------------------------------------------------------Youna.David s 81,522
FoUowine approval of the attencfina physician. eliaible subjects wiD be approached by our raan:b nurse. an imrestiptor, or an obstetrical resident or Jlafrphysician familiar with the INdy. lnlormalion pampblets wiD be circulated to physic:iln oftk:a and prenatal education classes to inform women or our racan:h. Our research protocol and consent form has been approved by the Human lnvaaiption Committee. FICUity of Medicine. Memorial University ofNewf'aundlaDd. and HaJth Care Corporuion of St. John's Resan:h Committee. .
To address in part iuua ofpneralizability and volumeer bias. demapaphic and au1come da&a wiU be coUected on parturients who are ineliiJible, or dcdine randomization. but allow confidenlial use of this information.
Wath the cxceprion of the experimental manoeuvre. all upec:u ofparient care will be determined by the patient's own physician. If eliaibility criteria are met and informed consent obcained, allocation wiU be carried out. Group allocation will be wiped by openina the next sequentially numbered opaque envelope only when induction is to bqin. Envelopes will contain a cud indic:a&ins Shldy drug packap to be adminislered, which was determined using a computer aenented random sequence aftd prepared by uniwnity research personnel not involved in the IIUdy. Randomization wiD be blocked in aroups or four and suatified by cervical ripeness (Bishop score <7, or not). Study dnsa pacbaes wiD be pRpiRCl by pharmacy and comain eiaht doses {maximum dose per subject used in our prior research has been seven) of a desipa~ed packet containina powdered misoprostol or methyl cellulose (visually indistin,Wsh3ble), to be adminiaterecl by oral syrinp after diaolvina in 30 cc of waaer, or in a 5 cc syrinae c:onuinina hydroxyethyl pi for vqinal use. For a aiven subject either the oral powder or the vaainal syrinp will concain Ktive misoprostol Our phlrmlcy has developed this approach bued on a previoully publilhed protocol'. The oral pnparation will be ldminisaered to a subject, immediately after the vqia11 syrinae has been emptied inlo the vap.al poslerior fornix. We have succeafidly used these adminiSIIUicms in a number of patienu. This approach has been necessary u the c:osa of a pllcebo llblet &om the phum : 'rical manufaclurer hu been quoaed u in eccess of$200,000.
A vaainal eum for cervical VI nent (Bilbap ICOI'e) and piiCiment of study medicalion (by a rllllt pllylician or raidenl) will occur immediately before randomizalion. Allsaudy induc:danl wiD be Cllried out oa ID inpltiem buis. with contiNaous electronic FHil and Ulerine CIDIIInCtion moai1orina unlil uterine ICiivity has resolved. or a mininum or 1 hour after lilY PG admilillrllion. Miloplosaollbldy drup wiD be npaaed at 4 hour intervals until one or the followinl occurs: proaaessiwlabar, contnctioa hquenc:y oftllree per 10 minutes. nonrassurina FHil tncina. or delivery. AD cllcilioRs with reprd to AR.OM. analpsia. epidural use. llld oxytoc:iD IUIIftlllll1ion will be mille by the lltlftdiaa physician. A alfphysiciu wiD ~ tbe palient WOre ach llllminillrllio of misoprosaol. The rationale for this schedule is our prior experience. aua This dose hqucncy will not be exceeded. Study dna& use !NilJ be clisconliiUed after Npture of membranes. Oxytocin aupenwion may not beain until 4 hours after dilconliradna misoprasaol. Intravenous oxytocin wiD beain at 2 miU/min and be inc:reued at 2 miU/min inca ements every 3().6() milwtes.
l2.7
133
Youaa. David S 8l,S2l
Mac..W ud newbom outcome ...uures are required. All study subjects will remain in dleir IDOCIIed poup for outcome Ill II iaeul • Ill "intent to trell" analysis.
The primary outcome ia the lime &om onset of induction to vaainal birth. Althouah con.nz may be abWned eulicr, ruadomizalion will not occur until the time induction is to 1Jeain. A di&renc:e of 240 minutes wu c:holcn u clinically imponant to detect, on the buis of a quellionnaire survey of prepant ud posq,utum pllients, nunins lnd physic:iu stafF.
Secondaly outcomes include:
I. Labor intervals to vqinal birth - induction to fi.dJ dilalion., duration of membrane rupture, llbor lllpl (fint. second and third).
2. Labor intervals with Caesarean binh.
3. Malcrnll incervencicms ud morbidity- frequency of IV use, Vlginal-exams (for study drup ud other}, oxytocin use, lllll1aia use (epidural, narcotic. general, other), nonreassurina FHR. fetal blood umplin& episiotomy, perineal laceration (depee}, incac:t perineum, manual ranoVII ofpllccnta, endometritis, fever >38 -con cwo oc:cuions more dian six houn apart. postp1rNm stay, estimaled peripanum blood loss. poscpartum haemofrhlae. blood transftasion. ocher sipificam illness and dCIIh.
4. Binb route: Vlainal {spontaneous, vacuum or forceps}, and Caesarean.
5. Frequency of uterine tacbysyscole (>6 coftCndions per 10 minutes).
6. Malemal Sllis&ction with labor CVIIUIIed by questionnaire (Labor Agentry Scale").
7. Neonatal morbidity- coni uteay ICid bue analysis, Appr scores ac one and five minutes, meconium prior to birth. ACOG criteria for binh IXphysia [four findinp: profound metabolic or mixed 1cidvnia (coni artery pH less thin 7.00), 5 mimlte Appr scare orJ or laa. neoaatal neuroloaic lbaonnllity,lnd dysf\aftc:lion of one other lnajar body sysaem}. time ill intlftlive care and halpitaJ. sepsis. pneumonia. mcninaitis. antibiotic treacmenL. other sipifiCIIII i1lnca or deldl.
1. MMimll G1 iatoleruot- hquency ofnausa, wmitina or clianhea by CU'IIIiWr NpOn ... pllieat questiollnlire.
Bllllinl demoplphic dala coUected on study subjects wiU include: indicalion for induction. ptrily, Bisbop scare. qe. liCe. heiaflt. weipt, poup 8 streptococcus (GBS) swus and neonllll bildlweiJfiL
12.1
134
s 11,522
The primary outcome measure rime &om induction to vapl&l binh. is a continuous vuilble.
• (2 tliled) • 0.05
p •0.20
z. •1.96
z, •0.84
ct • standard deviation derived fi'om poolin1 of results &om our prior vaainal miloprostol cohort •• (511 minutes) and n IIUoprostol intlct membranes lh'ltWil 21 (524 minutes).
N (group) • 2 ( (Z. +Z,)oiAtjl •16 (usina PEPI. Version 2, 1995. Computer Proarams for
Epidemioloaic Analysis. USD) 2N •172
Our prior SNdies •u• bad consistent overall Caesarean rues of 12.2 and 12. 4%. Combined. this rue is 12.3% with 99% CJ 1.8- 16.5%. Addifta 20% or 34 to the prior 2N would lllow for llllicipated Caesuan births in our sample.
:. RCT I Sample Size • 206 ,......., 1'111rw .. lppi'OXimacely 2600 annual binha II the Once General Hospital, SL John's,
Newla """""d . nil provincill tatilry uait is lbe oaly binh cen1• within 100 lcilomecen for a ....... ofov.l50,000. Our induclion rate is just over 20% Pvina approximalely 520 inducliana ..... ,.,. Our pnvioulstudia enrolled 222 and %19 subj«ts respectMIIy 1'
11 for induc:aioa with intact mea1ibr&Mi in 6 months each. It should be quite feuible to complete this clinical trill in one year, linaaltaneously with RCT D.
Dala AuiJiil
Hypolbllil teslina wiD be pcrb1aed only on the primary CMCOme. Other c:ompuilona willie eanli du.SIIypathelis aa•llion. Prelet1iaa decilion _. ud hypodlaes to be lellld lhauld aiwiaai::e cilia clredPIIIId post hoc sipificuce biu. Analylil wiD be on 111 .. innna 10 .._,.bail. t1le pnm.y outcome.......,., tm.10 vqinal birth. wiD be sipific:an& iCP<O.OS {U per ample a clfcrlluion) uliaa l*lllllllic llalialica (Student's t).
12.9
135
APPENDIX G: BUDGET
CJedentieh:
Dlali& • F...-u~-~~.· -"----":·--- .. ... • cau.:t _. callall1 . .. .... I8U I IUm l8d GUtliOIIIe dilL • laput~·-·;-. ,~.::- .. . .. • Jnfann p I lialjWIIdpnlllbaut nrilupi ....... PI a 0 .... !lin
illduc:ID ......... ..__ . • Palticiplle in iDfarllled caa.lll PI I • • Adminileer pallpUtulft •~'•& lien • r Ndre. • EdeM •tellllrlina IIIII ........, _,lUCia u &mily pbyD ·Ins, ..... ._,,
llld llbaur IDd *~~wry ........... - -.dy .. UIDUll. • Candnctlitnlule _._an ...._.IDpics. • Prepue ........... ie. llida.
Sllarill:
sJom •o
136
8r lit (25% Groll Slllly • CCP, PPP, UIC, WWC, Group. etc.) 2 woo
T .... ....,_ M9MQIO
Mataialud S.pp'n• ,.... ,_, CCIII ofNilaprollal.
Calk I ?'": S0.77 Jp.lienl X 790..... - s 601.00 GST(2.31%) • 14.04 PST (12.14%) • 72 M
$701.90
Calf..,_: SIO.OO /pllilnl x 790 pllilala • $7,900.00 GST (2.31%) • 112.49
PST (12.14% x CIDit ofpllcebo) - • o•• 36 S9,096.15
CalctdadMI: loral prior to..... - s 50.00 GST(2.31%) • 1.15
PST (12,14% XCIDitof'apendlbler fJi ·~ s 57.57
Ellimlle CIOil ofpllallllfa (S40fmal0).
Oladlldlll: Iaiii prior 10.... - s 410.00 GST (2.31%) • 11.09
PST (12.14% X COlt otapendlble) • 61 §] $552.72
l'rladJic: EJrimeted COlt ofphacac CJ?Jilla-
c.lc II'-~ 10.02/COfll X 4,000 capia • S 10.00 GST (2.31%) • 1.15
PST (12.14% X C1D1t af'prinlinl} • IQ 27 s 92.12
137
Trawl: Trawl Car ..-• f • of......_llld aplalbw fbtunl nllled ~ opportunitia.
£trimate trawl - s 2.000.00
Tetalper,_.. s 62,341.26
138